Electronics For You June 2017 3v614h

Nexte Issu

Contents nd Beyo om and Telec • 5G t and ring e s d an ctu H fa e u il n : • Mob vice Ma ards De er Bo tions elop a • Dev ial Applic str Indu

June 2017 Vol. 06 | No. 02 ISSN-2454-4426

Editor

: ramesh Chopra

Editorial : Editorial Secretary CorrESpondEnCE phone: 011-26810601; E-mail: [email protected] miSSing iSSuES baCk iSSuES, bookS, Cds, pCbs etc. nEwSStand diStribution

: kits‘n’Spares, new delhi phone: 011-26371661, 26371662 E-mail: [email protected]

mEntS : ph: 011-26810601 or 02 or 03 nEw dElhi (hEad offiCE) E-mail: [email protected] mumbai

Green Mobility Focus

54

GUIdES

92 BharatQR code: Not just for payments

26 E-PAYMENTS: Pitfalls in digital money

93 How about WhatsApp for marketing? 95 360-degree videos to lower costs

75 SOFTWARE: Open Source software you can use for IoT development

97 How Big Data analytics yields big gains

109 eStyle DIY: Turn your phone into a walkie-talkie

18 BIOTECH: Neuron-Like electronics

bEngaluru

: ph: 080-25260394, 25260023 E-mail: [email protected]

30 ARTIFICIAL INTELLIGENCE: The superpower brain

punE

: ph: 08800295610, 09870682995 E-mail: [email protected]

gujarat:

: ph: 079-61344948 E-mail: [email protected]

34 ARTIFICIAL INTELLIGENCE: A focus on image processing with neural networks

China

: power pioneer group inc. ph: (86 755) 83729797, (86) 13923802595 E-mail: [email protected]

japan

: tandem inc., ph: 81-3-3541-4166 E-mail: [email protected]

SingaporE

: publicitas Singapore pte ltd ph: +65-6836 2272 E-mail: [email protected]

taiwan

: j.k. media, ph: 886-2-87726780 ext. 10 E-mail: [email protected]

unitEd StatES

: E & tech media ph: +1 860 536 6677 E-mail: [email protected]

SUBSCRIPTION RATES Period Newstand Price Year (`) two 2400 one 1200

You Pay (`) 1800 960

Digital (`) 750 280

Overseas Digital Print uS$ 15 — uS$ 5 uS$ 120

please send payments only in favour of EFY Enterprises Pvt Ltd

4

June 2017 | electronics For You plus

110 eStyle BUYERS’ GUIDE: Wireless headphones cut the cord

Do-It-Yourself

40 TECH FOCUS: Artificial intelligence getting real, local 49 DESIGN: EDA becoming interactive with 3D rendering and routing 84 INNOVATION: HBand: Where your hand becomes the phone 86 INNOVATION: Livescribe, Neil, ili, LyfieEye, Kuri, U by Moen

printed, published and owned by ramesh Chopra. printed at international print-o-pack ltd, C-4 to C-11, hosiery Complex, phase-ii Extension, noida-201305, gautam budh nagar, uttar pradesh, on the first day of each month and published from d-87/1, okhla industrial area, phase-1, new delhi 110020. Copyright 2017. all rights reserved throughout the world. reproduction of any material from this magazine in any manner without the written permission of the publisher is prohibited. although every effort is made to ensure accuracy, no responsibility whatsoever is taken for any loss due to publishing errors. articles that cannot be used are returned to the authors if accompanied by a self-addressed and sufficiently stamped envelope. but no responsibility is taken for any loss or delay in returning the material. EfY will not be responsible for any wrong claims made by an r. disputes, if any, will be settled in a new delhi court only.

the World?

Smart World

rESEarCH & dEVEloPmENt

: ph: 022-24950047, 24928520 E-mail: [email protected]

What’s Happening Around

EVErYtHING YoU WaNtEd to KNoW aBoUt… 58 TEST & MEASUREMENT: Electronic diamond tester secrets and technology 64 AUTOMOTIVE: Self-driving cars platform and their trends (Part 2 of 2)

112 Auto-reversible DC motor with speed and rotation control 116 Ultra-low-power sensor hub using nRF24L01 modules 122 3W/6W audio amplifier using TDA2003 123 Dual-channel active AC analogue probe 125 Electronic tone generator system 127 PC-based oscilloscope using Arduino 129 Image compression using discrete cosine transform technique

NEWS

INdUStrY

12 98 105 107

78 LED LIGHTING: LED-filament bulbs: Growing your business with aesthetics

EFY Plus DVD

72 DATA INTERFACE: HSIC versus USB

88 INTERVIEW: Rameshwar Thakur, Chairman, TAL Manufacturing: Robots made in India cost lower… 90 INTERVIEW: Pradeep David, Universal Robots: Collaborative robots help create jobs… 101 MARKET SURVEY: Strategic Electronics Set for Revamp

Tech News Industry News New Products eStyle: First Look

Free tools to make your PCB smarter ................. 82 ‘Balance the books’ with FreeCAD ........................ I Pad2Pad: Manufacturer at your service................IV Lock your design with ZenitPCB suite ................VII Regulars 08 : Your suggestions 10 Q&A: Things you wanted to know

www.eFYmag.com

Your suggestions

3-PhAse InductIon MotoR In ‘Controlling 3-Phase Induction Motors Using VFD and PLC’ article published in April issue, I tried to control the speed of a stepper motor with Omron PLC and VFD but in vain. Please help. Muhammad Saleem Ullah Through email The author Joby Antony replies: The article talked about how 3-phase induction motors can be controlled using VFD and PLC. Stepper motors work altogether differently as these are pulse-driven, whereas induction motors are AC-driven. Therefore this project cannot be used directly for stepper motor control. However, since any PLC uses relays, one must connect each coil of the stepper motor through a relay for controlling it.

12V BAtteRy chARgeR I liked the ‘12V Battery Charger Cum Variable Power Supply’ DIY article published in November 2016 very much. Kindly add a voltmeter and ampere meter to the circuit. It would be of great help. Patrick Vaz Through email The author Fayaz Hassan replies: Voltmeter can be connected across point 1 and ground, and ammeter

From electronicsforu.com

Electronics Projects Can we use Arduino Uno in ‘Voice Controlled Home Automation System’ project published in October 2016 issue? Akshay EFY. Yes, you can use Arduino Uno but make sure you connect R1 to digital pin 6 of Arduino Uno, and pins TXD and RXD of HC-05 to RX0 and TX0 pins of Arduino Uno, respectively. The ‘0-50V Variable Power Supply Using LM317’ DIY project published in October 2016 issue is really very good. Ronaldo EFY. Thanks for the !

can be connected between point 1 and resistor R3. The modified circuit is shown here in the figure with the position of voltmeter marked as V and ammeter as A.

sound-oPeRAted tIMeR I assembled ‘Sound-Operated Timer’ DIY project published in January issue. When I clap in front of the mic, it gives output at pin 3 of NE555 (IC2), which doesn’t switch to another state as per the time delay set by timing components (VR2 and C4). Output remains on until the supply is turned off. I tried varying VR2 but nothing happens. Please help. Kunal Khosla Through email

Circuit of 12V battery charger with voltmeter and ammeter

8

June 2017 | electronics For You

EFY. First check potmeter VR2 and capacitor C4 for proper connections in the circuit. You may also try replacing capacitor C4 with any one of the values between 1µF and 10µF. Then observe the output after clapping in front of the mic. If you are using 6V at CON1, relay RL1 rating should be 6V.

Otherwise, the circuit will not work properly. You can also test the circuit by removing the relay section including diode D1. Then put a voltmeter across pin 3 of IC2. Check the output voltage before clapping and after clapping. See if the output goes off after a time delay.

Led As sensoR In ‘LED as Light Sensor’ DIY article published in December 2016 issue, the role of capacitor C1 is not clear. When I removed C1 from the circuit, the output remained the same. Pragya Rai Through email EFY. Here, capacitor C1 acts as an energy storage device. With capacitor C1, white LED2 will glow for some time even when the input is withdrawn. Whenever pin 6 of IC1 goes high, it starts charging and, at the same time, white LED2 glows. When pin 6 goes low, C1 starts discharging. White LED2 glows until C1 discharges completely. The circuit will work without C1 but that is not the main goal of the design.

MIcRoPython In ‘An Introduction To MicroPython’ article published in May issue, on page 45, under the subheading ‘Loading MicroPython on ESP8266 board,’ the first sentence “ESP8266 is a cheap Wi-Fi development board” is incorrect. ESP8266 is an IC, not a development board. Shashank T.S. Through email The author Ayan Pahwa replies: Depending on the ESP8266 board, it could be a development board or a WLAN module. www.eFYmag.com

Q&A

Things You WanTed To KnoW!

Ques. What kinds of pressure sensors are there in the market? pamarthi kanakaraja

ans. Pressure sensor is also termed as pressure transducer or pressure switch. Depending on the type of output interface, pressure sensors are classified as analogue and digital. A pressure sensor may also be called a different name (such as ‘load cell’ for a weighing machine) depending on the application. Different kinds of pressure sensors used in some popular applications are: 1. Industrial. Gases and pressures in industrial units, for example, in the chemical industry, are precisely monitored and controlled using pressure sensors. There are also board-level pressure sensors, normally mounted on PCBs, that are used to measure the pressure of gases or liquids. 2. Computing. Many computers and smartphones have touchscreen displays with pressure sensors. Whenever pressure is applied on the touchscreen, the sensor determines the point of application and accordingly generates an electric signal that informs the processor to take further actions. 3. Automotive. In automotive industry, pressure sensors monitor the oil and coolant pressure and regulate the power. Pressure sensors constitute part of the anti-lock braking system. Air bag systems also use pressure sensors to activate the bags during crash, ensuring safety of the vehicle occupants. 4. Medical. Pressure sensors are used in digital blood pressure monitors and ventilators to optimise these according to patient’s health and requirements. These 10

June 2017 | electronics For You

sensors utilise piezoresistive silicon MEMS technology. 5. Aviation. In airplanes, pressure sensors maintain a balance between the atmospheric pressure and control systems of the airplane. These not only protect the various internal components of the airplane but also give exact data about the external environment.

Fig. 1: Absolute gas pressure sensor

Fig. 2: Barometric pressure sensor

6. Marine. Pressure sensors are used in ships and submarines to estimate the depth at which these are operating and for detailing the marine conditions so that the electronic systems stay safe. Oxygen requirements are also regulated by pressure sensors. 7. Meteorology. Absolute pressure sensors measure pressure relative to a vacuum. Designed with a reference vacuum enclosed within the sensor itself, these can also measure atmospheric pressure. An absolute gas pressure sensor is shown in Fig. 1 and barometric pressure sensor in Fig. 2.

Q2. Why do 3G smartphones 4G netWork throuGh a Wi-fi router but not With a 4G sim card? pournima through e-mail

a2. There are two different combinations here: 3G phone with 4G SIM, and 3G phone with Wi-Fi router. 3G smartphones don’t 4G network because these work on wide-band code-division multiple access (WCDMA) technology. Whereas, 4G, the fourth generation of mobile telecommunica-

tions technology with the first release of Long Term Evolution (LTE) standard, is comparatively a new technology that is different from WCDMA. Therefore, theoretically, 4G SIM card will not work in a 3G handset. In order to use 4G SIM in a 3G smartphone, you need to enable LTE option. You can find many tips and tricks on the Internet claiming that 4G SIMs can be used in 3G smartphones. It is also claimed that a 4G SIM card in a 3G phone will give 3G speed only. However, some websites caution s that the trick may cause serious damage to their device and also may not work for all the devices. You can use 4G network on your 3G phone by connecting it to a portable Wi-Fi device running on 2.4GHz or 5GHz frequency because the phone is not directly accessing the service provider network, rather it’s the router

Fig. 3: 4G Wi-Fi router with 3G device

using the 4G network. The ‘3G’ of the phone is irrelevant, so it doesn’t even need a SIM card to access Wi-Fi network. However, the 4G portable WiFi device needs to use a 4G SIM card. The 3G phone, in turn, connects to the 4G portable Wi-Fi device through Wi-Fi tethering or mobile hotspot (Fig. 3). Answers compiled by Sani Theo, senior assistant editor (technical). Letters and questions for publication may be addressed to Editor, Electronics For You, D-87/1, Okhla Industrial Area, Phase 1, New Delhi 110020 (e-mail: [email protected]) and should include name and address of the sender www.eFYmag.com

Tech News Technology UpdaTes

Bionic hand that responds automatically A new generation of prosthetic limbs that allow the wearer to reach for objects automatically, without thinking—just like a real hand—are to be trialled for the first time. Led by biomedical engineers at Newcastle University and funded by the Engineering and Physical Sciences Research Council (EPSRC), the bionic hand is fitted with a camera which instantaneously takes a picture of the object in front of it,

Dr Kianoush Nazarpour demonstrates the bionic hand (Image courtesy: Mike Urwin/Newcastle University)

assesses its shape and size, and triggers a series of movements in the hand. Bying the usual processes which require the to see the object, physically stimulate the muscles in the arm and trigger a movement in the prosthetic limb, the hand ‘sees’ and reacts in one fluid movement. A small number of amputees have already trialled the new technology and now the Newcastle University team is working with experts to offer the ‘hands with eyes’ to patients at Newcastle’s Freeman Hospital. The findings of the study have been published in the Journal of Neural Engineering. “Current prosthetic hands are controlled via myoelectric signals—that is electrical activity of the muscles recorded from the skin surface of the stump. Controlling them takes practice, concentration and, crucially, time”—says Dr Kianoush Nazarpour, co-author on the study and a senior lecturer in biomedical engineering at Newcastle University. Using neural networks—the basis for artificial intelligence—lead author on the study Ghazal Ghazaei showed the computer numerous object images and taught it to recognise the ‘grip’ needed for different objects. This enables it to accurately assess and pick up an object which it has never seen before—a huge step forward in the development of bionic limbs.

Teaching robots to teach other robots Robots can now be made to do new types of tasks that they haven’t been able to learn before, like multistep assembly using both of their arms—thanks to C-LEARN approach developed by MIT researchers. Conventionally, most robots are programmed using one of two methods: learning from demonstration, in which they watch a task being done and then replicate it, or via motionplanning techniques such as optimisation or sampling, which require a programmer to explicitly specify a task’s goals and constraints. Both methods have drawbacks. Robots that learn from demonstration can’t easily transfer one skill they’ve learned to another situation and remain accurate. On the other hand, motion planning systems that use sampling or optimisation can adapt to these changes but are time-consuming, since they usually have to be hand-coded by expert programmers. 12

June 2017 | electronics For You

MIT doctoral candidate Claudia Pérez-D’Arpino discusses her work teaching the Optimus robot to perform various tasks, including picking up a bottle (Image courtesy: MIT CSAIL)

The new approach combines the intuitiveness of learning from demonstration with the precision of motion-planning algorithms: It allows non-coders to teach robots a range www.eFYmag.com

Tech News of tasks simply by providing some information about how objects are typically manipulated and then showing the robot a single demo of the task. Importantly, this enables s to teach robots skills that can be automatically transferred to other robots that have different ways of moving—a key timeand cost-saving measure for companies that want a range of robots to perform similar actions. The team tested the system on Optimus, a new two-armed robot designed for bomb disposal that they programmed to perform tasks such as opening doors, transporting objects, and extracting objects from containers. In simulations they showed that Optimus learned skills that could be seamlessly transferred to Atlas, a humanoid robot.

multifunctional printer has four nozzles to print the various specialised inks that make up the layers of the device—a base layer of silicone, top and bottom electrodes made of a conducting ink, a coil-shaped pressure sensor, and a sacrificial layer that holds the top layer in place while it sets. The ing sacrificial layer is later washed away in the final manufacturing process. Conventional 3D printing using liquid plastic is too hot and too rigid to use on the skin, while these flexible 3D printed sensors can stretch up to three times their original size.

3D printed skin to let robots feel

Researchers at Disney Research’s Wireless Systems group have devised an ultra-low-power system of sensors that transmit data to a central receiver by reflecting the ambient radio waves from commercial broadcasting systems. “Our idea is to reuse all the radio signals that are around us as a medium for transmitting data, much like sending ripples across a pond,” said Alanson Sample, associate lab director and team leader. This approach radically reduces the power requirements of the sensor nodes because it is the generation of radio waves that consumes most of their battery power. In a demonstration in a Disney Research lab, the researchers were able to meet the tiny bit of power demand that remained by using solar cells optimised for low-light conditions. Sample said the ultra-wide-band (UWB) approach—which backscatters all ambient sources—offers key advantages. Using multiple backscatter channels boosts the signal-to-noise ratio, substantially improving the sensitivity of the backscatter reader and decreasing dead zones. That, in turn, enables the system to operate on real-world ambient sources and substantially extends the range to 22 metres when using ambient signals from broadcast towers and 50 metres when using ambient signals generated by mobile phone up-link traffic. “The promise of the Internet of Things (IoT) is that wireless sensors will be ubiquitous, allowing devices to sense their environments and talk to each other,” said Markus Gross, vice president at Disney Research. “As we move towards connecting the next billion wireless devices to the Internet, however, the use of batteries to power these devices will become unworkable. UWB ambient backscatter systems, which potentially could be deployed in any metropolitan area, hold great potential for solving this dilemma.”

Researchers at the University of Minnesota have developed a revolutionary process for 3D printing stretchable electronic sensory devices that could give robots the ability to feel their environment.

3D printing stretchable electronic sensory devices that could give robots the ability to feel their environment (Image courtesy: http://odishatv.in)

“This stretchable electronic fabric we developed has many practical uses,” said Michael McAlpine, a University of Minnesota mechanical engineering associate professor and lead researcher on the study. “Putting this type of ‘bionic skin’ on surgical robots would give surgeons the ability to actually feel during minimally invasive surgeries, which would make surgery easier instead of just using cameras like they do now. These sensors could also make it easier for other robots to walk and interact with their environment.” This technology could eventually be used for health monitoring or by soldiers in the field to detect dangerous chemicals or explosives. “While we haven’t printed on human skin yet, we were able to print on the curved surface of a model hand using our technique,” McAlpine said. “We also interfaced a printed device with the skin and were surprised that the device was so sensitive that it could detect your pulse in real time.” McAlpine and his team made the sensing fabric with a one-of-a kind 3D printer they built in the lab. The 14

June 2017 | electronics For You

IoT sensors that connect via ambient radio waves

Tiny chip extends battery life of electronics Engineers at the University of Texas at San Antonio (UTSA) have developed a chip that can extend the battery life of lowpower electronic gadgets like cellphones. The chip is about www.eFYmag.com

Tech News

The chip from University of Texas at San Antonio (UTSA) makes batteries smaller and low-power electronics work more efficiently (Image courtesy: www.utsa.edu)

the size of a pin’s head, but has great potential in applications ranging from everyday electronics to medical devices. The chip can keep a phone working at top functionality with much less power. Moreover, it facilitates the use of smaller batteries, since the object itself is so small. The chip also tackles another common annoyance for electronics s: how hot devices get when they’re being used for several minutes. “The heat is a result of a lot of power being used,” said Ruyan Guo, team leader and professor of Electrical and Computer Engineering. “With our device there is less power consumption, which means the heat will be much less of an issue.

Heat-resistant capacitors Fraunhofer researchers have developed a capacitor that can withstand temperatures of up to 300 degrees Celsius. By comparison, conventional electronics can cope with temperatures only up to 125 degrees. Researchers did so by using an innovative mix of materials—and a special 3D trick. When manufacturing the conducting metal layers of the capacitor, the team etched tiny holes in the surface to increase its area. This 3D trick increases its capacity and simultaneously makes it possible to use a thicker dielectric. A thicker layer withstands higher temperatures better and can decrease uncontrolled leakage current in the capacitor. The experts adopted a new mix of materials in the production of the insulating dielectric too by using tantalum

Heat-resistant capacitor developed by Fraunhofer researchers can withstand temperatures of up to 300 degrees Celsius (Image courtesy: www.fraunhofer.de)

16

June 2017 | electronics For You

pentoxide—a compound of tantalum (a metal) and oxygen and aluminium oxide. Tantalum pentoxide is better at storing charge carriers than the silicon oxide usually used, so it increases the capacitor’s capacitance. In addition, the Fraunhofer researchers used a highly electrically conductive silicon as well as ruthenium, which is especially robust and heat-resistant. IoT systems usually link networks of sensors via radio, but radios demand battery power thus limiting usability. Disney Research has determined that one solution may be to get rid of the radios all together and communicate via the ambient radio waves from TV, radio and cellphones.

Search engine for 3D printable models What Google does for searching information on the Internet, new search engine IFind3D will do for printable 3D models. Launched by 3D Ninja, the largest reseller of 3D printers in The Netherlands, www.ifind3d.com contains over 700,000 designs. It combines online libraries for 3D printable models in one fast and responsive engine, making it easy for end s to find 3D printable models. At the moment, almost 70 per cent of all online libraries and repositories are connected. It is expected that before the year-end around 90 per cent of all worldwide 3D printable designs will be indexed in the search engine. IFind3D uses state-of-the-art algorithms to make sure the best results are served at all time using almost 100 different variables by integrating an AI service using IBM’s supercomputer ‘Watson.’

Pivotal technologies for IoT applications ON Semiconductor has developed a variety of flexible, energy-efficient, high-performance solutions for emerging Internet of Things (IoT) applications. The new AR0237 RGB-IR and AR0238 RGB-IR CMOS image sensors can capture daytime colour and nighttime nearinfra-red (nIR) image data on the same sensor without the cost and complexities (like refocusing and maintenance) of a mechanical IR-cut filter mounted onto the imaging assembly. Their unique colour filter array (CFA) arrangement employs a 4x4 kernel, which replaces some red and blue pixels with nIR-sensitive pixels and rearranges the spatial density of the remaining pixels. Thus, the devices can deal with the most challenging of lighting conditions. These devices are highly suited to home security and automated monitoring applications where ambient light levels vary greatly over time. From a connectivity perspective, the company has developed the RSL10 multi-protocol Bluetooth 5 certified radio system-on-chip. Claimed to offer the industry’s lowest power consumption in peak receiving and deep sleep mode, the RSL10 is optimised for use in IoT edge-node devices, as well as for health and wellness applications. www.eFYmag.com

biotech

NeuroN-Like Electronics

E Dr S.S. Verma is a professor at Department of Physics, Sant Longowal Institute of Engineering and Technology, Sangrur, Punjab

18

lectronics based human body and brain activities are miraculous and a great feat by the creator. It provides a lesson for the pioneers of science and technology to learn and mimic these activities by artificially creating electronically-controlled devices. The human brain is possibly the most complex entity in the Universe. It is absolutely remarkable and beautiful to contemplate, and the things you are capable of doing because of your brain are outstanding. The human brain is littered with some 100 billion nerve cells, together forming connections in tandem as each neuron is simultaneously engaged with another 1000 or so. In total, some 20 million billion calculations per second are performed by the brain. Inspired by the operation and structure of the brain, engineers and scientists are now developing bioinspired integrated circuit technology that mimics the neuron structure and operation of the human brain. Although processors have gotten smaller and faster over time, only a few computers can compete with the speed and computing power of the human brain and none comes close to the organ’s energy efficiency. So some engineers want to develop electronics that mimic how the brain computes to build more powerful and efficient devices. Scientists have developed a new prototype for electronic synapses to replicate the human brain, which could one day make neural networks incredibly clever. At present, neural networks built by research groups around the world consist of computers that are being trained using complex computer algorithms to solve complex problems and gain a deeper recognition and understanding of art and the world around us.

June 2017 | electronics For You

Memristors Memristors are the fourth class of electronic circuitry, alongside resistors, capacitors and inductors. Only confirmed to exist in 2008, memristors behave in a way similar to the synapses of neurons within the human brain. The resistance to current within a memristor is a product of currents that have previously flowed through it, meaning that the current flows easier as more current flows through. Due to these properties, memristors hold potential for non-volatile memories, and can make computers better at understanding speech, images and the world around these. In the past, it was very difficult to reproduce synapses because there are billions of neurons and thousands of synapses in the human brain, and the only way to get even remotely close to the power of a human brain using ordinary electronics is to utilise a gigantic amount of circuitry that would consume huge amounts of power. But then researchers managed to bring to life the concept of a memristor (using titanium oxide with two platinum electrodes), which means that conductivity in the memristor device changes depending on the charge ing through it, and connectivity can be changed from high to low. This is similar to the way brain synapses work, and could make it possible for much more energy-efficient computer systems to be developed that have memories that retain information even when the power is off. So in the future, computers would not need to boot up and could be switched off and on like an electric light. So now, researchers all over the world are trying to use memristors

www.eFYmag.com

biotech impulse of activity when a threshmake these more compatible with technology to develop electronic old is exceeded. A neuristor can current fabrication methods. synapses that work just like the be thought of as a slightly-leaky This new study does not deny human brain, to power neural netballoon that receives inputs in the long-standing efforts of the neuworks of the future, where computform of puffs of air. The only major romorphic community to build ers are capable of understanding difference is that more complex analogue neurons into chips. It and processing the information as neuristors can repeat the process is simply the next step towards well, and as quickly as a human again and again, as long as spikes getting people on board with being can. occur no faster than a certain reneuristor hardware instead of The brain’s neurons encode incharge period known as the refracneuristor simulation. formation in the patterns and timing tory period. In order to mimic the amazing of spikes of activity. That encoding A neuristor uses a relatively human brain, we need a chip and is hard to model using electronic simple electronic circuit to generate its constituting elements behaving hardware because most electronspikes. Incoming signals charge a like the constituting elements of the ics use binary (0 and 1) switches. capacitor that is placed in paralbrain—transistors for neurons. A However, researchers have comtransistor in some ways bined memristors and behaves like a synapse, capacitors in a way that acting as a signal gate. allows for the creation of When two neurons are in spiking output patterns. connection, electrochemiMemristors are decal reactions through vices made of materials neurotransmitters relay that behave as insulators specific signals. In a real until these are heated, at synapse, calcium ions inwhich point these act as duce chemical signaling. conductors. Researchers Transistors use oxygen paired a memristor and ions instead, engulfed in a capacitor in a paralan 80-nanometre-thick lel circuit and applied a layer of samarium-nickecurrent. As the voltage late crystal, which is the heated it, the memrisanalogue to the synapse tor behaved as a resistor channel. until it reached a critical Neuristors are brain-like computer chips (Image courtesy: idoc-pub.sitiosdesbloqueados.org) When a voltage is temperature; then it beapplied to the crystal, came a conductor. That oxygen ions slip through, changswitching allowed for full release lel with a memristor. When that ing the conductive properties of the of the energy stored in the capacihappens, charge builds up on the lattice and altering signal-relaying tor and, thus, mimicked the spiking capacitor by incoming spikes’ capabilities. Strength of the conbehaviour of neurons. discharges, and we have a spiking nection is based on the time delay The system, which the researchneuron that comprises just two in the electric signal fed into it. It ers termed a neuristor, is a very simelementary circuit elements. is the same way for real neurons plified model of neuron behaviour Details of trying to build neuristhat get stronger as these relay and produces a much more regular tors have been largely unglamormore signals. Exploiting unusual spiking pattern than a real neuron. ous. In fact, most people trying properties in modern materials, the They believe that using a different to create artificial brains with synaptic transistor could mark the memristor and a more complicated neuristor-like elements have been beginning of a new kind of artificial circuit could allow them to more content to do so by simulating intelligence—one embedded not in closely reproduce neuron behaviour these on regular old digital computsmart algorithms but in the very on a computer chip. ers. A group has come up with an architecture of a computer. improved method of fabricating a Neuristors Each time a neuron initiates an neuristor made from niobium-dioxaction and another neuron reacts, ide (NbO2). For now, their protoA neuristor is the simplest possible the synapse between these increasdevice that can capture the essentypes are too inefficient to be used es the strength of its connection; tial property of a neuron, that is, in large numbers on a single chip, the faster the neurons spike each the ability to generate a spike or but the group is developing ways to 20

June 2017 | electronics For You

www.eFYmag.com

biotech time, the stronger is the synaptic connection. Essentially, it memorises the action between neurons. So it does, in fact, run a bit like a neuron, in the sense that it adapts, strengthens and weakens connections according to external stimuli. Also, opposed to traditional transistors, the creation is not restricted to the binary system of 0s and 1s and, interestingly enough, runs on non-volatile memory, which means that, even when power is interrupted, the device re its state. Still, it cannot form new connections like a human neuron can. By exploiting the extreme sensitivity of this material, a very small excitation allows to get a large signal, so the input energy required to drive this switching is potentially very small. That could translate into a large boost for energy efficiency. It does have a significant advantage over the human brain—these transistors can run at high temperatures exceeding 160 degree Celsius. This kind of heat typically boils the human brain.

Technological developments The idea of building bio-inspired cognitive adaptive solid-state devices has been around for decades. It forms the basis for synaptic electronics, a field of research that aims to build artificial synaptic devices to emulate the computation performed by biological synapses. Synapses dominate the architecture of the brain and are responsible for massive parallelism, structural plasticity and robustness of the brain. These are also crucial for biological computations that are needed for perception and learning. Therefore a compact nano-electronic device emulating the functions and plasticity of biological synapses will be the most important building block of brain-inspired computational systems. Scientific interest has expanded 22

June 2017 | electronics For You

Electrical pulses convert a doped chalcogenide from an amorphous phase (pink) into a crystalline one (blue). Once a certain amount has been converted, conductance of the material suddenly jumps, mimicking the firing of a neuron (Image courtesy: cen.acs.org)

towards building electronic systems that mimic the ability of the human brain in performing energy-efficient and fault-tolerant computation in a compact space, and research activities in this field have been growing rapidly. Various efforts in neuromorphic engineering—a new interdisciplinary field that includes nanotechnologies—are being carried out, and the goal is to design artificial neural systems with physical architectures similar to biological nervous systems. In research findings, scientists have developed a carbon nanotube synapse with the elementary dynamic logic, learning and memory functions of a biological synapse. Research efforts on braininspired computing and synaptic electronics can be understood considering the inefficiency of conventional computational systems in solving complex problems. Researchers are focusing on developing synapse-like electronic

devices with directed attention to material systems that have been investigated for non-volatile memory technologies. Interconnection scheme of phase-change memory (PCM) synapses to reach ultra-high density and compactness of brain is being developed. In the crossbar array architecture, PCM synapses lie between post-spike and pre-spike electrodes, inspired by biological synapses formed between pre-synaptic and post-synaptic neurons. A broad spectrum of device systems with programmable conductance inspired by already existing device technologies, such as PCM, resistive-change memory, conductive bridge-type memory, ferroelectric switches, carbon nanotube devices and three terminal devices or field effect transistors based devices, have been explored. Using pectin. Memristors can provide a bridge for interfacing electronic circuits with nervous systems, moving us closer to the www.eFYmag.com

biotech realisation of a double-layer perceptron, an element that can perform classification functions after an appropriate learning procedure. The main difficulty the research team faced was in understanding the complex electrochemical interplay that is the basis for memristive behaviour, which would give them the means to control it. Researchers addressed this challenge by using commercial polymers and modifying their electrochemical properties at the macroscopic level. The most surprising result was that it was possible to check the electrochemical response of the device by changing the formulation of gels acting as polyelectrolytes, allowing for the study of ionic exchanges related to the biological object, which activates the electrochemical response of the conductive polymer. These developments open the way to make compatible polyaniline based devices with an interface that should be naturally-, biologicallyand electrochemically-compatible and functional. Pectin is a key ingredient for making delicious jellies and jams, not as a component for a complex hybrid device that links biological and electronic systems. But a team of scientists has built on previous work in this field using pectin with a high degree of methylation as the medium to create a new architecture of a hybrid device with a double-layered polyelectrolyte that alone drives memristive behaviour. The team of researchers has explained the creation of the hybrid device. In this research, they applied materials generally used in the pharmaceutical and food industries in electrochemical devices. The idea of using the buffering capability of these bio-compatible materials as solid polyelectrolyte is completely innovative, and it is the first time that these bio-polymers have been used in devices based on 24

June 2017 | electronics For You

organic polymers and in a memristive device. The next steps are interfacing the memristor network with other living beings, for example, plants and, ultimately, the realisation of hybrid systems that can learn and perform logic/classification functions. Using phase-change material. A nano-sized device made from a chalcogenide performs neuron-like calculations. Electrical pulses convert a doped chalcogenide from an amorphous phase into a crystalline one. Once a certain amount has been converted, conductance of the material suddenly jumps, mimicking the firing of a neuron. A research team has reported that nano-sized devices made from phase-change materials can mimic how neurons fire to perform certain calculations. This report shows quite concretely that we can make simple but effective hardware mimics of neurons, which could be made really small and therefore have low operating powers. The device imitates how an individual neuron integrates incoming signals from other neurons to determine when it should fire. These input signals change the electrical potential across the neuron’s membrane— some increase it, others decrease it. Once that potential es a certain threshold, the neuron fires. Previously, engineers have mimicked this process using combinations of capacitors and silicon transistors, which can be complex and difficult to scale down. The new work demonstrates a potentially simpler system that uses a phasechange material to play the part of a neuron’s membrane potential. The doped chalcogenide Ge2Sb2Te5, which has been tested in conventional memory devices, can exist in two phases: glassy amorphous and crystalline. Electrical pulses slowly convert the material from amorphous to crystalline,

which, in turn, changes its conductance. At a certain level of phase change, the material’s conductance suddenly jumps, and the device fires like a neuron.

Last words Different material systems or devices have strengths in various characteristics and required characteristics for a synaptic device that will enable the development of brain-inspired systems technology that scales to biological levels of device density, parallelism and functionality. The synaptic device is a simple twoterminal nano-scale device that can reach brain-level parallelism and compactness. Main characteristics are the ones that are directly related to parallelism, energy efficiency and fault tolerance. The synaptic device should emulate plasticity by implementing an analogue-like transition between different conductance states with very low energy consumption per synaptic event. Exact performance metrics depend on the target application and the scale of the system design. Although most of the work focuses on some specific applications for synaptic devices, the field has the potential to be utilised in a broad range of computing applications, especially the ones at the intersection of sensing and computation, where real-time and parallel processing of large-scale data is crucial. There is an enormous opportunity to completely rethink the design of computational systems in order to gain orders of magnitude of improvement in computational efficiency through inspiration from the biological brain. More interactions among different research disciplines—devices, circuits, architecture and computing—can further cultivate the synaptic electronics field and help define more targeted research paths for the future. www.eFYmag.com

e-payments

Pitfalls In Digital Money

S Manu Prasad is M.Tech in VLSI and embedded systems, and is currently working as assistant professor at AWH Engineering College, Kerala. His interests include VLSI, EDA tools, MATLAB and Latex

Fig. 1: ATM security measures

26

ince the government’s move to demonetise old ` 500 and ` 1000 notes last year, usage of digital money has drastically increased in the country. As India takes the cashless route and emerges as the biggest digital economy in the world, it also opens up many opportunities for cyber criminals. According to Norton Cyber Security insight report 2016, those who were not careful have experienced negative consequences including identity theft, money stolen from bank s, credit cards issued in their names and unauthorised apps installed on their devices. Data from Ministry of Finance in India shows that about 50 leading banks in India have lost around ` 4.85 billion between April 2013 and November 2016 because of cyber frauds. Fifty six per cent of this money was lost due to net banking thefts and credit/debit card cloning. Discussed in this article are some possible security threats and what you can do to avoid these.

your PIN or save it on your mobile phone. And, while entering your PIN at an ATM or while paying for stuff at a shop/restaurant, cover it using your hand or body so that no one else can view it. Also, change your PIN regularly. Use credit cards instead. For purchases, use credit cards instead of debit cards, because the former offer greater protection. You can also block credit cards at any time. Stick to own bank ATM. It is preferable to use your own bank’s ATMs instead of a different bank’s. This also saves you from paying transactions fees that you might have to pay after a certain number of transactions. Avoid saving card data. Most online vendors offer card saving options for easier and faster transactions in the future. Avoid doing so wherever you can. Do not let anyone else use you card. At shops or restaurants, always swipe your card yourself or have it swiped in front of your eyes. This is a good way for criminals to otherwise get access to all your details.

Debit/credit card fraud

Digital wallets

A debit or credit card fraud occurs when a criminal gains access to your debit/credit card details and the PIN to make purchases or withdraw money from your . To safeguard your card, follow the steps given below. Sign up for banking alerts. g up for banking alerts that are delivered as SMSes to your ed mobile number or email ID can help you keep track of your transactions. Safeguard your PIN. Never write down

Online wallets are easy targets. Many a time these do not have very advanced security measurements. This makes these vulnerable to attack. Many wallets need only one-time to carry out the transactions, and these keep you signed in. This helps hackers to access your easily. Wallets usually have a simple oneclick feature for easy operation. So if you lose your phone, anyone who lays hands on it can control your mobile wallet. Let us see what you can do to protect your digital wallet. 1. Install a basic antivirus and malware scanner on you mobile phone. 2. Check for the rating and credibility of the app before installing it. 3. Use a one-time (OTP) instead of a regular . 4. Log out of the mobile wallet after the transaction is complete.

June 2017 | electronics For You

www.eFYmag.com

e-payments rangeet.com/emoticons/eby.php?acct:personal

(a)

(b)

Fig. 2: (a) Original State bank website for Internet banking, and (b) mimic website example

5. Make sure the wallet provider has a customer wing that can help in case of fraud. 6. Do not use the same for all s. 7. Avoid installing third-party software shown on pop-up ments while using apps. 8. Double-check the details while transferring money. Even if the wallet has a possibility of threat, it is still safer than using a credit/debit card for online transactions.

Threats in Internet banking Even though banks provide a lot of security measures, online banking services (NEFT/RTGS) are not free from the menace. Security threats include phishing, spyware and adware, viruses, trojans and key-loggers, among others. Phishing and spam emails look like these have been sent from the bank itself, and ask for personal details such as names and s. Spyware and adware are a special type of software created by hackers to collect sensitive information. These track your Internet habits and interests, and provide the same to marketing companies. This is how you end up with spam in your inboxes. A virus attaches itself to a program by clubbing with another program in the computer by using different resources of the computer. An email virus is a special type 28

June 2017 | electronics For You

of virus that gets embedded in an email and when you open the email, it replicates your information and distributes it. A trojan can also be categorised as virus but it does not replicate itself and does not need any particular program to attack. It is a destructive program that acts as a harmless application, such as as helpful software claiming to get rid of unwanted software or virus. Instead, it attacks your information and makes it vulnerable. Key-loggers log all activities on the computer by capturing every keystroke. The sensitive information is then captured and accessed by fraudsters to attack s. Some security measures that you can take while using Internet banking are: 1. Always access the ed website of the bank. Do not follow any links sent in emails. 2. Beware of phishing attacks. Do not reply or disclose any sensitive information like name and in response to emails claiming to be from your bank. Banks never ask for sensitive details. 3. Protect your , and do not write it down. Always use strong s that are difficult to hack. Change your regularly—at least once a month. 4. Install a good antivirus on your system for protection from virus and spyware. 5. Check for https:// instead of

http:// in your browser address while using Internet banking. 6. Avoid using public computers or Internet café computers for accessing bank s. If you need to do so, use the onscreen keyboard for typing out the name and , check for unnecessary background programs, clear history and cache after logging out, and hide the while typing it out. 7. Do not click on unwanted ads on emails or other websites.

Mobile security A lot of us use phone banking these days, but most of us do not check the permissions we give while installing apps on our mobiles. Our mobile phones contain very sensitive information, and giving permission to access these means making it easier for criminals to hack our systems. Always be careful while accessing bank or financial data through mobile phones. 1. Do not use the same for wallets and bank s. 2. Use dynamic -generation apps for creating s for bank s. 3. Check while granting permissions when installing apps. 4. Avoid third-party software that pop up while using other apps. 5. Always log out after accessing your . 6. Do not save names and s on the app. www.eFYmag.com

artificial intelligence

The Superpower Brain

J Meghraj Singh Beniwal is B.Tech in electronics and communication, currently working as automation engineer at Infosys, Pune. He is a freelance writer and an Android app developer

Fig. 1: Evolution of artificial intelligence

ohn McCarthy explains artificial intelligence (AI) as “the science and engineering of making intelligent machines, particularly intellectual computer programs.” AI is a method to turn a computer into a computer-controlled robot or a software-controlled machine that ‘thinks’ wisely, in a manner similar to the way intelligent human beings think. It is done by studying humans—how they think, study, choose and work while trying to resolve a problem—and then by means of the outcomes of this study, developing appropriate software and systems. AI combines the characteristics of computer science, physiology and philosophy. It creates a machine that is unnaturally bright by following such programs and equipment that are able to take decisions on their own. Researchers are making systems that can emulate the thoughts of human beings, identify their speech and interact with them. AI is made up of two words: artificial and intelligence. Artificial. It is something that is not ordinary, but is made using human skills. It implies creating a replica or simulation of a human brain. Though you can make a machine unnaturally related to human beings, it lacks naturalness and spontaneity.

Interesting facts about AI AI started in 1950. Let us discuss some AI facts that are crucial in today’s era. AI can repair itself. A recently-developed robot can rebuild itself even after trailing two of its six legs. The robot does not know what is broken, but it can notice the performance being dropped. Therefore using an algorithm, it can determine what is wrong and how to amend it. AI can write. Data that is captured from seismographs is used by a computer to write an article, which then turns it into figures and plugs those figures into a beautiful story. It takes the graphs and compiles these into well-written reports.

Intelligence. This means injecting intelligence into a machine so that it can easily carry out the work that would otherwise need a human brain. The machine can make its own decisions, according to the situation.

How AI originated The growth of AI started with the desire of introducing intelligence in machines that was similar to the intelligence in human beings. AI has the following two basic goals: Creating expert systems. Systems that exhibit intelligent behaviour and can study, reveal, describe and advise its s. Implementing human intelligence in machines. To make systems that comprehend, think, study and act like human beings. British mathematician Alan Mathison Turing, known as the father of AI, predicted in the early 1950s that, going ahead machines would perform human acts that even human beings would fail to achieve. He devised Turing Test to test a machine’s ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human being.

Real-life applications of AI AI has an enormous effect on our daily life, and its power is likely to rise in the 30

June 2017 | electronics For You

www.eFYmag.com

artificial intelligence coming years. Let us explore some examples of AI that we are using in our day-to-day life. Transportation. AI has already been granted a licence for driving cars on the roads of California, USA. Routine work. Robots can do mundane work like run household errands, supervise routine ing and reports compiling, manage vehicles and so on. And, now these are capable of creating art, too. Marketing. Rocket Fuel, a marketing tool, uses analytical science for

Philosophy

Computer Science Psychology

Sociology

Artificial Intelligence

Maths

Neuron Science

Biology

Fig. 2: Components of AI

adapting and scaling marketing return on investment by means of media and CRM channels. By considering diverse factors and data combinations, the engine bids on the required media and optimises campaigns.

AI tools Companies like Google, Facebook and Amazon are investing a lot of money in their R&D devisions, and are buying startups that work in areas like machine learning, natural language and image processing. A research at Stanford University has concluded that gradually more useful applications of AI, with thoughtful, positive impacts on our culture and economy, are expected to come out soon. Let us take a look at some of the well-known AI tools. Personal. Apple’s Siri, Microsoft’s Cortana and Google Now are 32

June 2017 | electronics For You

intelligent personal assistants that use AI. Business. Examples of AI tools used for businesses are: 1. Gluru helps systematise online documents and emails, and uses AI to get the latest actionable information. 2. x.ai lets AI synchronise schedules and acts as a personal scheduler. 3. CrystalKnows incorporates AI to help you understand the best way to communicate with others. Developers. For developers, some AI tools are: 1. Vicarious helps build the next generation of AI algorithms. 2. Prediction.io has easy-to-employ, open templates for a selection of advanced AI workloads. 3. h2o.ai helps build smarter AI applications that are rapid and scalable.

Limitations of AI The main disadvantage of AI is the cost incurred in its protection and repair. All programs have to be efficient enough to suit the changing needs, and machines need to be made smarter. Some of us have concerns regarding the application of AI, as they feel it is wrong to install AI into a machine so as to make it work for their benefit. Some of the reasons given by them are: If robots start to substitute human beings in every field, it will ultimately lead to unemployment. With so much help from machines, if human beings stop using their thinking abilities, these abilities will steadily decline. With AI, they may turn out to be overly dependent on machines, thereby losing their mental capabilities. If control of the machines goes in wrong hands, it may cause a havoc. Apart from these, there is the fear of robots overruling humans. Practically, we should be the masters of machines. But if things turn the other way round, clever machines could turn out to be smarter than us. www.eFYmag.com

artificial intelligence Machine Vision

A Focus On Image ProcessIng With Neural Networks

S Saurabh Durgapal is working as technology journalist at EFY

ome time ago there were some very interesting product releases from the likes of Google and Microsoft. While Google announced a new assistant in Google Allo, which is quite friendly, Microsoft announced a puck-like device for artists working on Surface Pro devices. Google’s chat assistant on my phone read my personal messages and gave suggestions in reply. Is this just a matter of word association, or an invasion of privacy? However, my intrigue began when a friend sent an image of his presence at a club. Allo showed some automatic reply options including, “Have a good time!” To me, it seemed like a fun and convenient way to reduce the hassle of typing out a full message.

Analysing images is a serious task Photography is one area that employs image analytics heavily. Dr Joseph Reger, chief technology officer, Fujitsu EMEIA, explains, “Modern cameras identify faces to make focusing easier, or these can identify smiling faces or faces of pre-selected people.” Image analysis starts by first identifying edges and moves on to full shapes. However, making changes in certain key areas in the images has successfully fooled such systems. More on that later, first let us look Artificial neural networks process information the way biological nervous systems like the brain do, for optimal decision making. Certain characteristics of their architecture and the way these process information makes these superior to conventional techniques on a certain class of optimisation problems. Optimisation techniques are called into play every day for industrial planning, allocation, scheduling and decision-making, among others. So how do you optimise the network?  Select a network architecture, that is, number of hidden layers, number of neurons in each layer and activation function.  Initialise weights randomly (weights being the mathematical formula).  Use forward propagation to determine the output node.  Find the error of the model using known labels.  Back-propagate the error into the network and determine the error for each node.  Update the weights to minimise gradient.

34

June 2017 | electronics For You

at how a computer recognises and processes images. Image analysis ranges from finding basic shapes, detecting edges, removing noise and counting objects, to finding anomalies in the routine. Security and surveillance are incomplete without an intelligent system that can detect pedestrians and vehicles, and make a difference. So how does image analytics work? How was Google’s new assistant able to analyse images and suggest possible replies as soon as I received the image? Uncanny Vision, a startup focused on image analysis, helps clarify how it happens. Their system identifies objects in images using an analytical system powered by deep learning and smart algorithms. The deep learning framework, Caffe, powers this system. The framework managed by Berkeley Vision and Learning Center (BVLC) is capable of processing more than 60 million images in a single day using just one NVIDIA K40 GPU. On an average, this amounts to about 695 images per second, whereas we only need 17 to make a series of images into a video, with HD video going up to 60. Basically, this system can analyse about 11 HD videos at once. Deep learning is affecting image analytics. Deep learning is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, linear and non-linear. It is a representational learning approach ideally suited for image analysis challenges ranging from security and dailypurpose applications to medical problems. Object detection is considered to be the most basic application of computer vision. Enhancements on top of object detection result in other developments including anomaly mapping and error detection in real-time imaging. However, deep learning www.eFYmag.com

artificial intelligence a smaller tree, it means that the model did not capture underlying trends properly, and the tree needs to be re-examined. Divide and conquer to analyse. Cluster analysis, on the other hand, categorises objects in test data into differUpgrading neural ent groups or clusters. Test data networks can be grouped into clusters based on any number of paIt is no doubt that neural networks can identify and recogThe Scream by Edvard Munch on the left, and the horrifying scream rameters, resulting in multiple after exaggeration of patterns on the right algorithms in cluster analysis. nise patterns and do a lot of Clusters are formed based on other interesting stuff. However, Chinese game of Go is a complex your algorithm, making it a multiwhen we talk about real-time image board game requiring intuition, objective optimisation problem. analysis from multiple angles and creative and strategic thinking. It Formation of clusters, however, is lack of content in the frame, going has been a major subject in artificial not successful with the current data beyond the capabilities of neural intelligence (AI) research. In March analysis needs. With the introducnetworks is required. One of the 2016, Google’s AlphaGo beat Lee tion of the Internet of Things (IoT), most stated advancement in this Sedol, a South Korean professional data that needs to be analysed has regard has been convolutional neural Go player of ‘9 dan’ rank by 4-1. reached much higher volume. Many networks (CNNs). Self-driving cars, And here we thought recognising methods fail due to the curse of auto-tagging of friends in pictures, dimensionality, resulting in many pafacial-security features, gesture recog- and following a person in a crowd was amazing for a neural network. rameters being left out while optimisnition and automatic number plate ing the algorithm. recognition are some of the areas Lesser-known image analytics If you can remove some pathat benefit from CNNs. rameters in clustering, removing it Broadly, a CNN consists of three Image analysis can also be done altogether is not a stretch. Dimension steps that can be divided into more using many other techniques. These reduction is the process of reducing steps. Starting with convolution, we can be as simple as a binary tree the number of random variables bego back to our engineering days, with a simple true-false decision, or ing considered. The set of principal when we convolved a filter with input as complex as structured prediction. variables is then processed through function to get some output. Padding Decision tree, for example, is used feature selection and extraction. was then added to make this result by Microsoft Structured Query LanData is classified by filters based on similar in size to the original image. guage (SQL) server. Never thought the features to be added or removed You know the basics. we would be using analytics in SQL, while building the model based on Pooling is done to reduce the size did we? But when you think about prediction errors. of data, and this makes it easier to sift how data is fetched in SQL, it makes Feature extraction then transthrough. Then, the networks use fula lot more sense. forms this high-dimensional data into ly-connected layers where each pixel Decision tree is a predictive modfewer dimensions through principal is considered as a separate neuron elling approach used in statistics, component analysis, among other similar to a regular neural network. data mining and machine learning. techniques. Application areas include Batch normalisation was anAn analogy for easier understanding neuroscience and searching on live other step in the process, which has would be the binary tree. Some of video streams, among others, as rebecome outdated with use. Better the popular implementations are IBM moval of multi-collinearity improves results over CNNs result in similar SPSS Modeler, RapidMiner, Microsoft the learning system. implementation as well. SQL Server, MATLAB and the proOdd man out algorithms. Used Where we have seen CNN before. gramming language R, among others. mostly in data mining, anomaly detecVideo analysis, image recognition In R, decision tree uses a comtion is another interesting algorithm. and drug discovery are some of the plexity parameter to measure tradeAnomalies, also referred to as outliers, common uses of CNNs around us. offs between model complexity and novelties, noise, deviations and excepHowever, the one that brought CNNs accuracy on training set. A smaller tions, can be a fancier version of the to light would be AlphaGo procomplexity parameter leads to a bigodd man out. Typical anomalies ingramme by Google DeepMind. The ger tree, and vice versa. If you have is still an expanding topic. Dr Reger adds, “It is still unclear what the optimal depth of a deep learning net should be, or what the ideal component functions in the targeted approximation might be.”

36

June 2017 | electronics For You

www.eFYmag.com

biotech artificial intelligence implanted routine cathclude bank during frauds, astructural defects, eterisation procedure. It requires medical problems or errors in text. no leads or implanted batteries, Image analysis by Uncanny Vision is andexample. sends data a secure Web the an Theto system analyses platform. The sensor is and designed images through a CCTV points to improve thewith quality of life and out anomalies, basics going medical as well as lower down to aoutcomes, person falling. costs for those who suffer from

• www.lulu.com • www.magzter.com • Createspace.com • www.readwhere.com

32 38

February | electronics For you June 20172017 | electronics For You

congestive heartreleasing failure and other Yonatan Zunger, an apolcardiac ailments.to rectify the error, ogy and attempts Golden Gopher Magnetic resulting in the removal of the tag BiosensingThe system. This portable, altogether. explanation behind low-cost and easy-to-use device the error came to light once it was detects biomarkers released for public. in human serum andThe urine, as well as heavy problem with Googlemetals AI was in lake water. Knownimages. as Z-lab,Trainthis training using animal Proper hardware highly-sensitive sensor now in a ing an AI is a process toisdetermine is also necessary development for smaller, large set of parameters thatmore cali-portable models. brates a function that maps image Neural networks have existed since dataElectronic to content.Aspirin. Results For maypeople vary the early 1990s, but initially success who suffer migraines, based on thefrom images used in clustrainwas very limited due to the reter headaches andimages other causes of ing. Some sample tested by stricted availability of hardware and chronic, excruciating facial enthusiasts gave somehead very or creepy inadequate learning methods. “Last pain, the ‘take two aspirins and call results. Even a relatively simple decade has been a welcome change me in network the morning’ method is useneural over-interpreted an in this regard,” says Dr Reger. He less. Doctors have long associated image, resulting in trippy images. adds, “Advanced learning algorithms theEngineers most severe, chronic research forms of at Google’s and massive 16-bit FP performance Fig.modern 5: Endotronix wireless health headache sphenopalatine lab decidedwith to check on patternsganin hardware havemonitoring turned glion (SPG), a facial nerve bundle, the system recognised and set up neural networks into an effective havethat not tweaks yet found treatment abut system theapatterns technology for image analysis.” that has a long-term effect to exaggerate the result. An on im-SPG. Movidius Myriad 2, Eyeriss and Electronic Aspirin is a patientage ed through enough times Microsoft HoloLens are some vipoweredintool blocking SPG radiresulted thefor image changing sion microprocessors being used in signals at basically the first sign of a some headcally, and tripping today’s vision-processing systems. ache, and involves the permanent classy paintings. These may include direct interfaces implant of a small nerve-stimulating to take data from cameras with a How deviceitinlooks the upper gum on the side greater emphasis on on-chip dataof the head normally affected flow. These are different than regu“After logic programming with by headache. Lead tip of the implant lar video-processing units as these languages like Prolog and expert Fig. 6: Golden Gopher Magnetic Biosensing system connectsthe with the SPG bundle, are suited more to running machine systems, present phase of ad-and when aneural patientnetworks senses the onset vision algorithms such as CNNs and vanced constitutes of athird headache, heAI,” or she scale-invariant feature transform the wave of saysplaces Dr Reger. a handheld remote controller on a (SIFT). Some of the application But we are still decades away from the cheek the implant. areas for vision processing include system thatnearest can gettoanywhere close Resulting signals stimulate SPG of robotics, the IoT, digital cameras for to a realistic one. This brilliance nerves and body blockispain-causing virtual reality and augmented realour human the reason why neurotransmitters. ity, and smartcameras. researchers have been trying to Google smart of lenses. break the enigma computer vision Fooling an AI Fig. 7: Electronic Aspirin Google has created smart of by analysing the visual mechanics lenses for people human beings or made other animals. Reports released last year suggested who suffer diabetes Background clutter,from hidden parts that changing some pixels in a those who simof images, and difference in illuminaphoto of an elephant could fool a ply wear in glasses. The tion and variations viewpoint neural net into thinking it is a car. technology is engineered are some areas that require focus in AI systems have committed some Soft lens to take thewe tears a development before canin move disturbing tasks as well. Aprilencapsulates Taylor electronics person’s eye and measure from iTech Post has explainedSensor in andetects glucose in tearstowards developing Jarvis. Accordglucose levels.networks For people ing to Dr Reger, “Neural in article how her dogs were categoChip and antenna receive power and wear glasses, general andwho image analysis inthe parrised as horses. send information would be of engineered ticular willlens remain some the most Twitter , jackyalcine, has what the companies interesting to and important topics in also reported some funny busisayinformation is ‘to restore the eye’s cutting-edge technology ness with Google AI. This has lead natural autofocus.’ in years to come.” to Google’s chief social architect, Fig. 8: Google’s smart lenses www.eFymag.com www.eFYmag.com

tech focus

ArtificiAl intelligence Getting Real, Local

A Janani Gopalakrishnan Vikram is a technically-qualified freelance writer, editor and handson mom based in Chennai

rtificial intelligence (AI) is undoubtedly one of the hottest verticals to watch in 2017. An integral part of gaming and robotics, now it has also deeply penetrated aspects of the real world that influence everybody. As the core driving force of analytics, AI is at the heart of every vertical from industrial operations and banking to security and the network of devices. It has also become part and parcel of consumer electronics, as was evident from this year’s CES 2017, and an inseparable part of social media as we can see from the new features being launched by Facebook, Google and other Internet majors. The increased usage of AI technologies fuels further growth in the space—not just in the form of research but also new products such as chips for artificial intelligence. The arena is also teeming with start-ups that are bringing fresh new ideas to life. Some see AI as a threat to their jobs, while others see it as an opportunity to do better. Either way, it is not something you can ignore. So, well, we try to give you a glimpse of what is happening here...

AI shakes up the chip market The word ‘smart’ has become like a compulsory prefix for most object names, from pens to phones and musical instruments.

Deep learning—the most obvious trend in AI today Deep learning is basically a nature-inspired technology that uses neural networks modelled after the human brain, to help a system learn by itself, just like a human child does. It is made of multi-layered deep neural networks that mimic the activity of the layers of neurons in the neocortex. Each layer tries to understand something more than the previous layer, thereby developing a deeper understanding of things. The resulting system is self-learning, which means that it is not restricted by what it has been taught to do. It can react according to the situation and even make decisions by itself. Deep learning usually requires large memory banks and runs on huge servers powered by advanced graphics processing units. However, recent innovations have opened up the possibility of implementing deep learning at the network edge. Find out more about deep learning: Deep Learning Makes Conventional Machine Learning Look Dumb (EFY Jan. 2017) 40

June 2017 | electronics For You

The moment the word ‘smart’ is used, it means some amount of intelligence (obviously artificial) is involved. This intelligence is usually implemented using large servers with built-in AI models based on neural networks. In general, the operation of neural networks can be understood in two stages. The first stage involves training the AI system to perform a given task. The second stage is the inference stage, which involves execution of AI models against live data. The training stage involves crunching immensely large data sets, to help the system identify and learn from patterns in the data. Data collected by the device flies across the Internet to data centres packed with servers, where this is done. On servers, central processing units (Us) are usually supplemented with other types of processors like graphical processing units (GPUs). Nvidia is one of the most well-known GPU makers, who has been powering many data centres for years. Last year, it launched Tesla P100—a GPU with an architecture focused on accelerating AI and deep learning. Apparently, Nvidia spent more than $2 billion on R&D to produce the new chip, which has a total of 15 billion transistors—around thrice as many as the company’s previous chips. According to a press release, an artificial neural network powered by this chip can learn from incoming data twelve times faster than Nvidia’s previous best chip. To show that machine learning requires a special architecture, Google also developed its own processor called the tensor processing unit (TPU). On the other hand, Microsoft went with field-programmable gate arrays (FPGAs) for its AI needs. Not to be left out of the race, Intel acquired Altera—a company that sells FPGAs to Microsoft. Together with Nervana Systems, a company it acquired last year, www.eFYmag.com

tech focus be used in remote farms or even by children playing hide-and-seek! EnRoute, another drone maker, has used on-board AI to help its drones navigate and fly faster, avoiding objects on their path. Cisco has developed a collaboration device that uses AI to recognise people in a room and automatically pick a field-of-view (FOV) with people in it instead of empty chairs. The FOV is spontaneously adjusted as people walk in and out, or move around. The system also zooms in on people who are speaking. Live Planet’s new 360-degree 3D camera for live streaming of video uses on-board AI to encode 3D videos in real time. Live Planet’s chief strategy office Khayyam Wakil explains, “The camera produces a stream of 65 gigabytes, which is too much data to transmit to a cloud server. On-board processing has made the live streaming AI goes local possible.” Sniffing the trend, Intel In a press event held in San Nvidia Jetson TX2 credit-card sized platform for intelligent edge acquired Movidius in 2016. Fransisco in March this year, devices like robots, drones, cameras and portable medical devices Movidius produces special(Courtesy: Nvidia) Deepu Talla, vice president ised low-power processor and general manager of chips for computer vision and deep Michigan, which incorporates a Nvidia’s Tegra business unit, cited learning. Their button-sized Myriad deep-learning processor capable of four reasons in favour of bringing 2 platform has many features that operating a neural network using AI technology to the edge (on implementation of deep just 288 microwatts. board the device)—bandwidth, learning at the network edge. Last year, Nvidia launched Drive latency, privacy and availability. As Myriad’s SHAVE processor engines PX2—a palm-sized platform to imthe number of devices communiachieve the hundreds of giga-flops plement auto cruise capabilities in cating with the cloud rises by the required in fundamental matrix mulautomobiles. This open AI car platday, there will be a dire shortage tiplication compute that is essential form features a unified architecture of bandwidth in the near future. for deep learning networks. The that allows deep neural networks to Latency is also an issue in applicaon-chip RAM keeps huge volumes of be trained on a system in the data tions like self-driving cars where intermediate data on the chip itself centre, and then deployed in the car. a split second’s delay can have to avoid bandwidth bottlenecks. The This year, Nvidia launched Jetson serious implications. Privacy, of platform comes with native TX2—a credit-card sized, plug-in course, has always been a serious for mixed precision and hardware edge-processing platform designed issue. So, when data is processed flexibility—both 16-bit and 32-bit for embedded computing. Teal within the device itself, there is no floating-point data types, as well as Drones has used the Jetson module lingering doubt about misuse on u8 and unorm8 types are ed. to develop a smart drone that can the cloud. Availability of the cloud The company literature explains understand and react to what its is also questionable in rural areas that existing hardware accelerators cameras are seeing. Since this drone where connectivity is unreliable. can be easily repurposed to provide does not rely on the cloud, it can When AI is implemented at the Intel is also building Lake Crest (or Nervana Engine), a full-stack solution for deep learning, optimised at every stage to make training ten times faster. The second stage, or the inference, is undergoing a major shakeup. In the past, most of this work was also done on the cloud using GPUs, TPUs and FPGAs. However, now smart device makers are beginning to feel that this is not ‘real-time’ enough. So, the trend is turning in favour of implementing the inference or execution stage at the edge (that is, in the device itself) instead of relying on the cloud. Since existing devices don’t have enough processing power, memory size and bandwidth for this, the semiconductor industry is facing a heavy demand for high-performance, low-power inference engines for deep neural networks, which can be built into devices.

42

June 2017 | electronics For You

edge, all these issues are sorted out. “We will see AI transferring to the edge,” Talla said to the press, with future intelligent applications using a combination of edge and cloud processing. This requirement to build intelligence into the device is leading to a major bustle in the semiconductor industry, not to forget a lot of hardware innovation. In last month’s story on smart robotics, we read about a micromote (a chip measuring just one cubic millimetre) developed at the University of

www.eFYmag.com

tech focus Excerpts from Chia-Chiunn Ho’s recollections of Larry Zitnick’s AI class at Facebook  Researchers like Zitnick were building machine vision one tiny piece at time, applying very particular techniques to very particular parts of the problem. But then academics like Geoff Hinton showed that a single piece—a deep neural network—could achieve far more. Rather than code a system by hand, Hinton and company built neural networks that could learn tasks largely on their own by analysing vast amounts of data. “We saw this huge step change with deep learning,” Zitnick says. “Things started to work.”  A neural net mimics the web of neurons in the brain. It operates by sending information between processing units, or nodes, that stand in for neurons. But these nodes are really just linear algebra and calculus that can identify patterns in data.  Each layer makes a calculation and es it to the next. Then, using a technique called ‘back propagation,’ the layers send information back down the chain as a means of error correction. As the years went by and technology advanced, neural networks could train on much larger amounts of data using much larger amounts of computing power.  “For the first time ever, we could take raw input data like audio and images and make sense of them,” Zitnick told his class.

the flexibility needed to achieve high performance for convolution computation. Myriad also comes with a development kit that includes dedicated software libraries to sustained performance on matrix multiplication and multidimensional convolution. Start-up Graphcore proposes to handle deep learning with a socalled intelligent processing unit (IPU)—a graph processor that can manage both training and inference on the same architecture, and eventually across multiple form factors (server and device) too. The chip is expected to get ready for early usage by year-end. According to the company, “This same architecture can be designed to suit both training and inference. In some cases, you can design a piece of hardware that can be used for training, then segment that up or virtualise it in a way to many different s for inference or even different machine learning model deployments. There will be cases when everything is embedded, for instance, and you need a slightly different implementation, but it’s the same hardware architecture. That’s our thesis—one architecture, the IPU, for training, inference, and 44

June 2017 | electronics For You

 Convolutional neural networks is a method inspired by the brain’s visual cortex that groups neurons into ‘receptive fields’ arranged almost like overlapping tiles.  Long short-term memory networks (LSTMs) are neural networks that include their own short-term memory—a way of retaining one piece of information while examining what comes next. This is what helps identify the commands you speak into Android phones.  In the end, all these methods are still just math. But to understand how they work, students must visualise how they operate across time (as data es through the neural network) and space (as those tile-like receptive fields examine each section of a photo). Applying these methods to real problems, as Zitnick’s students do during the workshops, is a process of trial, error and intuition— kind of like manning the mixing console in a recording studio. You’re not at a physical console. You’re at a laptop, sending commands to machines in Facebook data centres across the Internet, where the neural networks do their training. But you spend your time adjusting all sorts of virtual knobs—the size of the dataset, the speed of the training, the relative influence of each node—until you get the right mix.

different implementations of that machine that can be used in servers, cloud, or at the edge of the network.” That would be the ultimate thing to wish for!

AI seems so real At one time, the term ‘AI’ was associated only with robots, but now it is everywhere—from security cameras to cars and enterprise applications. In Pittsburg, USA, AI is helping solve traffic woes. Speaking at a White House Frontiers Conference,

Carnegie Mellon University professor of robotics Stephen Smith said that traffic congestion costs the U.S. economy $121 billion a year, mostly due to lost productivity, and produces about 25 billion kilograms of carbon dioxide emissions. The AI-based smart traffic management system piloted in the city has reduced travel time by 25 per cent, idling time by over 40 per cent and emissions by 21 per cent. Unlike conventional traffic lights that have pre-programmed

Surtrac is an intelligent approach to traffic management, implemented in Pittsburgh, USA www.eFYmag.com

tech focus When logic wins over machine learning When we drive, we are faced with unexpected potential dangers— people crossing the road, dogs that don’t obey their masters, children riding cycles and more. However, from experience we learn to ignore some and pay more attention to others. But most autonomous cars try to handle each situation theoretically, which is too tiresome for the car and its occupants! nuTonomy, a company that has been testing its robo-taxis in Singapore’s One-North technology business district, uses formal logic to handle such complex situations. Formal logic is based on a hierarchy of rules, with priorities. For example, the rule “don’t hit pedestrians” has a greater priority than “don’t hit other vehicles,” which is more important than “don’t hit objects.” The car attempts to apply all the rules, but in the given order of priority. It uses a planning algorithm called RRT* to evaluate many potential paths based on the data from cameras and other sensors. Decision-making software evaluates these paths and selects the one that fits in with the rule hierarchy. The idea of formal logic is different from machine learning. In machine learning, you show the algorithm innumerable driving scenarios, from which it figures out the rules of good driving and then applies them to real-life situations. nuTonomy uses machine learning to interpret sensor data but not for decision-making because the company feels that machine learning is like a ‘black box’ and it is difficult to figure out why machine-learning systems make the choices

timings, the Surtrac system applies AI algorithms to data collected by the radar sensors and cameras of computerised traffic lights to dynamically build a timing plan. The system is decentralised and each signal makes its own timing decision. It also sends the data to traffic intersections downstream so they can plan ahead. There are 50 such smart intersections now, with plans for expansion citywide. Following that, Smith’s group wants to improve the system to enable signals to talk to cars! According to an IEEE news report, they have already installed short-range radios at 24 intersections. Such systems are expected to begin being built into some cars this year. Traffic signals can then let drivers know of traffic conditions or change in lights, increasing safety and relieving congestion. The vehicle-to-infrastructure communication system could also prioritise certain vehicles like public transport buses. AI is helpful on social media too. Facebook, for instance, 46

June 2017 | electronics For You

nuTonomy relies on formal logic for their robo-taxis to make the right decisions (Courtesy: nuTonomy)

they do. Formal logic, on the other hand, gives more control over the machine and guarantees that the car will obey rules in the order of importance and stay safe even in unprecedented situations. “It’s a rigorous algorithmic process that’s translating specifications on how the car should behave into verifiable software,” explains nuTonomy CEO and cofounder Karl Iganemma, in a press report. “That’s something that’s really been lacking in the industry.”

uses AI to spot and remove offensive content. It is also planning to integrate AI-based suicide prevention tools into Facebook Live and Messenger, in order to recognise and help people with suicidal tendencies.

Bosch-backed robot Kuri (Courtesy: Mayfield Robotics)

In April this year, Mark Zuckerberg unveiled a platform that transforms s’ smartphone camera into an engine for augmented reality (AR). The solution relies on implementing AI on the network edge. The platform lets s to layer digital effects atop images and videos captured by the camera. One of the fun demos showed digital sharks swimming around a bowl of cereal. Facebook has more real-world plans for the future. For example, you can pin a virtual note on your fridge and your roommates will be able to see it when they view the fridge through their cameras. Neural networks help to identify people and track their movement and activities within the camera’s FOV, in order to apply appropriate digital effects. Facebook’s deep neural networks run on the phone itself because getting across the Internet and back will be too slow to effectively implement such digital effects. While Facebook has optimised its deep learning technology to run on current-day mobile phones, www.eFYmag.com

tech focus they feel things are bound to get has become real now, and fiction difficult as digital effects get more today is being chiselled into reality complex. But, Facebook expects at labs across the world—and by that the future hardware enhancestart-ups too. ments will surely boost their So far, AI has been achieved machine learning models. mainly using complex algorithms. CES 2017 was full of AI-powered Now, imagine a chip that by itself consumer products. Apart from the works like a synapse of the human expected dominance of AI-enabled smartphones, wearable devices, home appliances and cars, two interesting developments relate to operating systems and home assistants. According to industry experts, more than 40 million homes will have a home assistant by 2021. In November last year, Google launched Google Home for this segment. Amazon’s Alexa is too well-known to be introduced again. Samsung has come up with Otto, and Bosch is backing Kuri. Facebook too demonstrated a personal assistant, though there is no informaTwenty Two Motors’ smart scooter for Indian roads tion about its availability (Courtesy: Twenty Two Motors) yet. Apple is also supposedly working on a Siri-powered brain—wouldn’t it make AI more home assistant. real and human-like than ever beWith so many connected and fore? Researchers at CNRS, Thales, intelligent devices all around us, have managed to create directly on people are getting very worried a chip an artificial synapse that is about privacy and data safety. So, capable of learning. You can use companies like Google and Norton these chips to create intelligent have also come up with solutions to systems comprising a network of secure your devices. Google is offersynapses, requiring much less time ing Android Things—an operating and energy. system that powers smart devices Another system developed at the and the Internet of Things (IoT) in Sandia National Laboratories aims a secure way. Norton Core is a moto improve the accuracy with which bile-enabled Wi-Fi router equipped cybersecurity threats (or bad apples) with machine learning and Symanare detected. The brain-inspired tec’s threat intelligence techniques Neuromorphic Cyber Microscope to defend your home network from designed by the lab can look for potential threats. complex patterns that indicate specific bad apples, consuming Lots more on the anvil less power than a standard 60-watt light bulb. This small processor was Researchers all over the world are found to be more than a hundred still exploring the possibilities of times faster and a thousand times AI. What was fiction a decade ago www.eFYmag.com

more energy-efficient than racks of conventional cybersecurity systems. Lots of interesting AI research is happening at MIT too. Last month, MIT researchers presented a paper proposing a fast and inexpensive way to achieve speech recognition. Current speech recognition systems require a computer to analyse innumerable audio files and their transcriptions, to understand which acoustic features correspond to which typed words. However, providing these transcripts to the machine learning system is a costly and time-consuming affair. This limits speech recognition to a small number of languages. The new approach proposed by the researchers does not rely on transcripts. Instead, the system analyses the correlation between images and spoken descriptions of those images, as captured in a large collection of audio recordings. It eventually learns which acoustic features of the recordings correlate with which image characteristics. According to the scientists, this is more natural—more like the way humans learn. Plus, it is less expensive and less time-consuming, opening up the possibility of extending speech recognition to a larger number of languages. Another research at MIT attempts to make machines predict the future. For humans, it is easy to understand what will happen when a player bowls a ball or when a car skids off the road. However, it is not so easy for machines. A team at MIT is developing a deep-learning algorithm for this. From a still image of a scene, the system can create a short video simulating the future of that scene. Researchers feel that such generative videos can be used to add animation to still images, detect anomalies in security footage, compress data for storing and sending longer videos, and so on. electronics For You | June 2017

47

tech focus nuTonomy prefers formal logic to machine learning because the latter is a black box, with no way to find out why the system made a particular decision. This is also one of the reasons why many people in the medical profession shun computerised decisions. MIT might have the answer in store. They have figured out a way of training neural networks so that these not only provide predictions and classifications but also the rationale behind each decision. In their research, the team divided the neural net into two modules. The first module extracts segments of text from the training data, then assigns scores to segments according to their length and coherence. The shorter the segment, and the more of it drawn from strings of consecutive words, the higher the segment score. The segments selected by the first module are then ed on to the second module, which performs the prediction or classification task. The modules are trained together, with the aim to maximise both the score of the extracted segments and the accuracy of prediction or classification. The researchers have successfully validated this technique on many textual data sets such as reviews on a website, freeform questions and answers, and pathology reports. As confidence in AI increases, we find it being used in critical applications too—aviation being an example. A team of researchers at the University College, London, is developing a new AI-based Intelligent Autopilot System that learns how to manage emergencies by watching how well-trained pilots do so, and then applying this learning to similar situations. The system is trained like professional pilots. According to a press release, the team uses a high-fidelity, professional version of the desktop flight simulator X-Plane to teach the autopilot 48

June 2017 | electronics For You

to fly a Boeing 777, subjecting it to severe weather conditions, engine failures, fires and emergency landings or turnarounds. Closer home, AI-powered scooters may hit the road by 2018. Haryana-based start-up Twenty Two Motors has developed a prototype smart scooter that is powered by AI and connected to the cloud. An app allows the to remotely control and access the scooter, while the cloud system helps in automatic troubleshooting. The scooter applies AI to data collected by sensors on the scooter, to understand the rider’s behaviour. It also enables decisions like the best route to take depending on battery conditions, target location and topography of the route (presence of bridges, flyovers, etc). The start-up raised ` 100 million in April this year and plans to launch the scooter at next year’s Auto Expo. Not a self-driving scooter but definitely smart enough to begin with! There is a lot of software-based AI innovation happening too, like the one by Mumbai-based Arya.ai, which offers deep learning algorithms for developers to build intelligent systems that can learn, adapt and do things with minimal inputs from humans. The DL Studio platform can be used to incorporate intelligence in e-commerce platforms, diagnostic assistants, image processors for drones, security, device management and maintenance, and more. Deep neural networks are also scalable. Another interesting trend is the availability of different AI capabilities as services that can be quickly deployed in applications. Clarifai’s powerful visual recognition application programming interface (API) is one example. It uses machine learning to automatically tag, organise and search visual content. Similarly, Datalog.ai offers conversational intelligence as a service for virtual assistants, bots, devices and corporate applications.

For developers, it is as easy as plug-and-play. No complex infrastructure or development is needed to put AI to work.

Building trust AI is indeed at an inflection point. What has made it so hot today? While some would credit the availability of powerful computers or advances in statistical machine learning and deep learning techniques, others say that AI has attained this level of focus and investment mainly because of the sheer amount of data that the IoT is churning up. With sensors all around us, networks are dizzy with data flying all around. Somebody sitting on data obviously wants to make sense of it. So, there is a greater demand for AI, and demand always drives supply—that is the underlying principle of commerce. The industry is bustling to meet this demand and the air is rife with partnerships and acquisitions. Still, there is one big challenge in the way of deploying AI in real-world scenarios: You need to win the trust of people before they accept AI as a way of life. There is a lot of doubt about security and privacy. The industry is coming together to solve this bottleneck. Last year, Amazon, DeepMind/Google, Facebook, IBM and Microsoft announced the formation of a non-profit organisation called Partnership on AI to improve public understanding of AI technologies and formulate best practices for development and deployment of AI. The IEEE Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems is also an effort towards aligning the development of AI and autonomous systems with the values of its s and society. As long as such basic ethical requirements are met and we are assured that intelligent devices will not overthrow us, artificial intelligence is definitely hard to resist! www.eFYmag.com

design

EDA Becoming IntErActIvE With 3D rEnDErIng And routIng

E Saurabh Durgapal is working as technology journalist at EFY

lectronic design automation (EDA) is an indispensable tool for optimised circuits. Most engineers have worked on one such tool or the other. Marco-Casale Rossi, product marketing manager, design group, Synopsys Inc., explains the advances in EDA, “Today, we can floor-plan hundreds of millions of library instances, or we can place and route tens of millions of library instances in less than 24 hours.” “We have moved from deg 350nm to 10nm,” says Ruchir Dixit, technical director, Mentor Graphics. “There was a time when a four-layered board was among the most complex circuit designs. This has become the basic now.” We also have changes in the type of components being used and signal speeds, hence calling for efficient designs.

The ever increasing tools Since its emergence in the late 1970s, EDA has quickly evolved into a dedicated industry, and new players keep emerging at regular intervals. Today, we have advanced EDA tools like Altium Designer, LabVIEW and Virtuoso, among others, that help engineers come up with the best blueprint for complex circuits. Besides the industry-established tools, we also have some Cloud based tools that have revolutionised the way beginners and hobbyists take up electronics. Cloud based EDA tools like EasyEDA and MeowCAD allow them the ease and freedom to work with the latest libraries, while providing updated versions of the software, without the hassle of updating software. www.efymag.com

Cloud based tools. Imagine working on a tablet; you do not have to carry your heavy laptop everywhere. You do not have to think about storage, appropriate graphics and processor requirements. All you need is a browser and you are good to go. Along with traditional circuit simulation features, Cloud based tools provide ubiquitous access. This allows you to travel light with only the concern for the Internet speed (which is nothing new). Using various advances in website deg, these are now accessible over a wide variety of mobile platforms including smartphones and tablets. Cloud based tools also feature quick help via dedicated forums, which provide access to the community of like-minded individuals working on their own problems. Sharing your project with the developer community and your co-workers over the Cloud, without having to care for compatibility of software, has also come as an advantage with Cloud based EDA tools. Flip side to the Cloud. However, Cloud based tools are yet to establish themselves for high-precision circuit design. These are more in tune with hobbyists working on their individual projects, or students learning to work on EDA. But for professional purposes, system based tools remain the better choice. “An efficient tool should be in a system. This removes dependence on the Internet while working,” says Sreekanth Tamanna, country manager for India, Pakistan, Bangladesh and Sri Lanka, Altium Ltd. Processing time in a Cloud based tool has also increased. If you are working in a closed network, you have to often confirm you have the latest version in your designs. Improved performance, more complex deg, local storage and better working are some advantages with electronics for you | June 2017

49

design on-system EDA tools. This might sound like the best alternative, but let us consider other alternatives before judging a tool.

Opening up the source Open source E-CAD tools are becoming popular. With latest releases like Ki-CAD version 4.2, these have become more attractive than ever. With 3D rendering, Github library integration and for making SPICE simulation, these are attracting designers big time. Anuj Deshpande, founder, Makerville Solutions, says, “We can now make artistic circuits.” He adds, “We can write scripts in Python and convert to a desirable format using Ki-CAD.” Such features certainly make open source tools attractive. Let us look into some of these features that have cropped up in recent years. On the designers’ requirement, Neel Desai, product marketing manager, Lynx, strategic alliances and GTS marketing, Synopsys Inc., says, “Design managers need tools that can help them manage complex system-onchip (SoC) design projects, monitor their progress and forecast their outcomes.” Cadence design team has come up with massivelyparallel architecture, physically-aware context generation, unified global routing and PPA optimisation in their systems. This involves coordinating the design efforts of multiple engineers and analysing massive amounts of data, to achieve the fastest and most predictable tape-out possible. Some recent features are discussed next. 3D visualisation. Imagine making a circuit and being able to work on real-time problems, before having to face those problems in real time. Now, stop imagining and go for the latest tools and see for yourself. This option to visualise the full module before possibly destroying copper comes in very handy at times. “Parameters like heat dissipation can be calculated in 3D render itself,” says Tamanna. 3D models also help in checking the final design of the device. With a simple 3D render, the design can be modified to suit various design requirements. Power-distribution management. With the latest tools we can measure functioning parameters. This helps in checking if the design would work as expected or not. “If the board is getting heated in a particular area, then it has to be designed again,” says Tamanna. This can also help in preventing disaster from striking at a later stage on copper. Higher precision. With the reducing size of electronics on a daily basis, requirement for higher precision is also a concern. Earlier people were not concerned with a few millimetres. “But today, designers require very high precision,” says Tamanna. Today, we 50

June 2017 | electronics for you

www.efymag.com

design It is certifiably golden A small sub-set of EDA tools, which stands a class apart, are classified as golden. Also known as sign-off quality, these tools are highly accurate and ease the end-to-end process from RTL to GDS in case of ASIC development and timing closure for fieldprogrammable gate arrays. Several semiconductor foundries provide pre-evaluated reference methodologies, also called RM flows, consisting of suitable tools, versions and scripts in order to automate the entire design process. Ruchir Dixit, technical director, Mentor Graphics, explains some of these golden features as, “We have a more accurate implementation, and we check for those implementations.” The tools provide an indication for problems, which can then be handled. With added features, for how to fix those problems and how quickly can you fix it can be taken care of. He adds, “The system also s systems with multiple PCBs and provide a larger design environment.”

have requirements for precision at about 0.1mm. The latest EDA tools this requirement in order to bring about the best in design. Digital electronics designs also require very high precision. Current solutions are prepared to the best in design. “Composite EM technology in ADS platform maintains the same level of accuracy, and we have increased the performance by a large margin,” says Deepak R.K., general manager, EEs of EDA, Keysight Technologies India Pvt Ltd. Special cases of board design. Wearables and other special use cases have increased the demand for flexible PCBs. “Rigid-flex is a technology for designers working on flexible PCBs,” says Tamanna. With this new feature, designers can parameters that might interfere with different movements and shapes of the product. Some parameters to be kept in mind while deg flexible PCBs would be to not compromise on the physical and electrical integrity of the circuit. “Beyond certain parameters set while deg flexible PCBs, the board loses its integrity,” explains Dixit. Open source tools are also adding such features since wearables often employ flexible PCBs. Working in groups. Dixit says, “Designs have become complex. Calibre provides provisions for breaking the complex board into individual components.” These individual components can then be used by different people. Today, tools offer live collaboration for multiple people working in real time. Open source tools are also updating to the latest features with the group working feature being one. “The master-slave mode is for teams working on different levels of the project,” says Deshpande. This can also be used to monitor the progress of the design. Improving the finer points. Making the interface attractive and easy to use certainly helps, and the industry www.efymag.com

electronics for you | June 2017

51

design Major contributors to this report

Anuj Deshpande, founder, Makerville Solutions

Chris Allsup, technical marketing manager, senior staff, RTL synthesis and test automation, Synopsys Inc.

Marco-Casale Rossi, product marketing manager, design group, Synopsys Inc.

has been working in this direction as well. Deepak says, “In the last year, we have focused on making our tools more -friendly.” With the large number of available components and minute changes from one component to the next, accurate documentation is a necessity. Hence, “Efficient documentation tools are becoming a necessity,” says Tamanna. Updated tools have automatic schematic and PCB generation, which reduces errors due to human intervention.

Taking on the challenges Certain challenges can occur during deg of the circuit, mask preparation, testing the circuit and so on. For circuit deg, Rossi explains, “Once the objective is within reach, we hold placement and routing, and systematically change the gates—same footprint, different timing, power, temperature inversion point.” Routing the circuits. Nowadays, it is practically impossible to do custom/manual routing and, hence, designers use EDA tools to do auto-routing. According to Cadence design team, routing can take anywhere from a few hours to days, depending on the complexity of the circuit. For such complex SoCs that have billions of transistors, extracting resistance and capacitance is a Herculean task. Commercially-available field solvers do things automatically once placement and routing is done during design phase. Masking. Primary challenge in 52

June 2017 | electronics for you

Neel Desai, product marketing manager, Lynx, strategic alliances and GTS marketing, Synopsys Inc.

Ruchir Dixit, technical director, Mentor Graphics

mask creation is for semiconductor manufacturers to continue scaling at and below 16nm/14nm using 193nm ArFi lithography equipment. “Until next-generation lithography equipment is available for production, manufacturers must find techniques to print smaller and smaller geometries with existing equipment,” says Tom Ferry, vice president - marketing, silicon engineering group, Synopsys Inc. He adds, “Computational techniques modify the design layout to create mask designs that enable lithography equipment to produce patterns on masks that accurately represent the layout (designer intent).” Verification and synthesis. For many years, designers exclusively relied on stuck-at automatic test pattern generation to test ICs. Because delay faults caused by opens and shorts occur more frequently than stuck-at faults, they gradually began to employ transition delay automatic test pattern generation to improve defect coverage and lower defective parts per million. According to Chris Allsup, technical marketing manager, senior staff, RTL synthesis and test automation, Synopsys Inc., “At advanced nodes, process variations give rise to increasingly-subtle physical defects that require additional patterns that target more complex failure mechanisms such as partial shorts and partial opens.” The RTL netlist in Verilog format is now accompanied by a power

Sreekanth Tamanna, country manager for India, Pakistan, Bangladesh and Sri Lanka, Altium Ltd

Tom Ferry, vice president marketing, silicon engineering group, Synopsys Inc.

intent description in IEEE 1801 UPF format, and logic synthesis has been upgraded to understand power and ground, and to perform optimisations, taking into increasingly-complex concepts such as power/voltage/shutdown islands. Database issues. Compatibility among EDA tools is another parameter to be considered. While files from one tool should be able to work with other systems for designers to have maximum ease of transition, the tools have to be lightweight in order to work efficiently. Rossi puts to rest the database debate with, “There are no simple databases in EDA, but rather efficient binary dumps on disks of in-memory data representation/infrastructure of different classes of tools.”

Where the evolution is going “Today, there is no solution for continuity between our logic synthesis and our place-and-route tools,” adds Rossi. “Moving forward, the border between logic synthesis and place-and-route will blur, and possibly disappear.” Chip design is a constant battle against rising process and functional complexity on the one hand, and reducing time to market, on the other. Additions like the 5G library for system design, according to Deepak, helps innovation in technology, as well. We can hopefully look forward to a design environment that includes validation of IPs, libraries and foundry technology data at the beginning of the process. www.efymag.com

test & measurement

ElEctronic DiamonD tEstEr Secrets And Technology

G T.K. Hareendran is founder and promoter of TechNode Protolabz

emstone identification can be an arduous problem, and getting it right is essential for the people in the trade. Recently, electronic gem testers have come in the market that promise to simplify this problem. These are small, handheld devices with a small probe that is placed against the gemstone. These devices first entered the market as diamond testers, and helped solve the problem of distinguishing natural diamond from cubic zirconia (a synthetic diamond stimulant). An electronic diamond tester is, in fact, a portable device that we can put in our pocket. The tester has a small, needlelike tip that needs to be placed firmly on the stone under test. If the diamond is real, the device indicates that on its display, often with an aural alert.

How it works Fig. 1: Testing a diamond

58

Diamond is an extremely good conductor of heat, and this property has been the basis of

June 2017 | electronics For You

one of the oldest tests for diamond—touch a diamond with your lip and it will feel cold! Diamond testers’ working principle is based on the fact that different gemstones conduct heat differently. Heat will through a diamond in a different way than it would through a cubic zirconia or glass. The diamond tester will detect the rate at which heat moves through the stone and tell whether it is a real diamond or not. As diamond is an exceptional thermal conductor (and cubic zirconia is not), the first diamond testers used a heated probe to detect heat dissipation. These were very effective at distinguishing natural diamond from cubic zirconia. But then another diamond simulant called moissanite entered the market, and diamond testers were not able to detect it (moissanite is also a very good thermal conductor). Moissanite is a naturally-occurring mineral that was first discovered more than 100 years ago. At first glance, it looks very similar to diamond, and these two stones also seem to have the same physical properties. Although many people think of it as a substitute for diamond, moissanite’s chemical composition is quite different—diamond is made up of carbon, whereas moissanite is a form of silicon carbide. This recant called for improved testers that use electrical conductivity to test stones. Latest diamond testers on the market include both thermal and electrical conductivity test modes. The advantage here is that you only need one device instead of two, and it costs slightly less to buy the two-in-one tester/multi-tester (heat for diamond and electricity for moissanite) rather than two separate testers. Type II diamond tester. Moissanite testers work by measuring electrical conductivity through the stone (diamond is not electrically-conductive, and moissanite is). However, there is a very rare type of diamond (type II diamond) that has an unusuwww.eFYmag.com

test & measurement al chemical composition that makes it electrically conductive. So it will show as moissanite on a two-in-one tester/multi-tester. Fortunately, type II diamond testers are now available but cost a few extra bucks.

Role of UV light Some diamond testers (Fig. 3) have a built-in ultraviolet (UV) light source, and this has led to the logical assumption that UV light can be used for testing diamonds. Note that, there is absolutely no way we

can distinguish a diamond from a non-diamond using UV light. UV light does, however, have some use for professional gemologists because it can give an indication of probability when comparing a natural diamond with a synthetic one. And, UV light makes a difference when testing a moissanite using a moissanite tester or two-in-one tester/multi-tester. Moissanite testers that do not work on some odd moissanites work perfectly when the stone is exposed to UV light. This is because, according to the laws of Quantum Physics, UV light can make a noticeable difference to electrical conductivity.

Gem tester electronics Fig. 2: An electronic diamond tester

Fig. 3: Diamond tester with built-in UV light source

Fig. 4 shows the reference system diagram of a gem tester that uses high voltage to quickly test a gemstone for legitimacy. This basic concept is adaptable for determining a gem type based on its electrical conductivity. In principle, an electronic circuit (including the gem under test) as part of a circuit path is used to measure its electrical conductivity and, hence, the gem type. The procedure of determining whether a gem is moissanite or synthetic diamond involves providing a high voltage across a gem surface, greater

than a breakdown voltage, and measuring a minuscule current that flows through the gem. Here, the first (probe) and second (clip) s of the circuit couple the high voltage to the gem under test, while a low-impedance detector circuit is used to flag when the s are erroneously ing each other during measurement. The amount of current flowing through the device is measured using a high-impedance resistor referenced to ground. Voltage on the resistor is measured and compared to a predetermined threshold voltage. Measurements above the threshold voltage indicate that the gem is conductive, and that the gem therefore is a synthetic diamond or moissanite. Conversely, measurements below a threshold voltage indicate that practically no current flows through the gem, and that the gem therefore may be a diamond—if other tests such as a thermal conductivity test show it to be as one. At the brain of the gem tester is a microcontroller for processing a multitude of analogue inputs to produce a number of outputs for ensuring whether a measurement for a gem under test is, in fact, within a specific range, signaling that it is moissanite or synthetic diamond.

Getting the feel Battery

VCC

Switch

DC-DC Boost Conv.

DC-DC Buck Conv.

V Bat Monitor

VCC

HV Monitor

Visual Display

Microcontroller Probe

Clip

Low-Imp DET High-Imp DET

Fig. 4: System diagram of a gem tester

60

June 2017 | electronics For You

Audio Alert

Using a diamond tester/moissanite tester/multi-tester is not complicated. A good start is to read its guide/instruction manual line by line, and make sure you are not doing anything wrong. If your tester works consistently (gives the same reading on the same stone every time) and correctly (accurately identifies stones), then you do not need to know the principle on which it works. However, if you are having problems, then the following basic tips will help solve most problems: 1. The probe of the tester must be clean, especially if it has not www.eFYmag.com

test & measurement been used for some time. Clean it by rubbing it at right angle in a circular motion on a piece of white bond paper. 2. Clean the stone by wiping it with a clean cloth soaked in alcohol. Alcohol works best because it evaporates quickly and leaves no residue. 3. Place the tester probe on the stone at right angle, and press firmly to make good (hold the tip at right angle so as to not break it). You should get a result in about three seconds. If not, remove the probe, wait for a few seconds for the stone to cool down and try again. 4. Do not test the same stone repeatedly, until it becomes hot. Hot stones do not as diamond, even if these are. (Inconsistent readings miraculously become consistent if you wait a minute or two for the stone to cool down.) Further, relatively-cold stones give a high reaction on the LED indi-

62

June 2017 | electronics For You

Fig. 5: Electronic diamond tester with LED indicators

cator segments, and relatively-warm stones give a low and often sluggish reaction. (A very tiny diamond on a hot day may only move one or two LED segments up.) A multi-tester can show you if the stone is a diamond or not. It carries out the diamond test and then, a fraction of a second later, it carries out the moissanite test. Its microcontroller combines both readings and calculates a response. You will not get correct readings if, in that first split second, you remove and replace the probe or slide it across the stone or press hard, then let go

and again press hard. It only works if you put the probe in the centre of the stone firmly once. As with all testers, this is not a problem if you can see what you are doing and you have a steady hand. An electronic diamond tester with state-of-the-art technology assures quick, accurate readings to determine if a diamond is real or fake. Most multi-testers utilise a combination of the proven thermal and electrical conductivity principles. The tester probes, together with the electronics circuitries, are designed to pick up and segregate data collected from the stones via a customised microcontroller. Within a split second, the test result is displayed on an electronic display. However, no tester can offer 100 per cent guarantee of identifying all diamonds correctly. There are always exceptions, and these apply to almost all brands of electronic diamond testers.

www.eFYmag.com

AUTOMOTIVE Part 2 of 2

Self-Driving CarS Platform And Their Trends

I V.P. Sampath is a senior member of IEEE and a member of Institution of Engineers India. He is currently working as technical architect at AdeptChips, Bengaluru. He is a regular contributor to national newspapers, IEEE-MAS section, and has published international papers on VLSI and networks

64

n the first part of this article we covered SENSORIS innovation platform, and the steps needed to make a self-driving car possible. In this part we shall cover lidar and the rest. Lidar, which stands for light detection and ranging, consists of a cone- or puck-shaped device that projects lasers that bounce off objects to create a high-resolution map of the environment in real time. In addition to helping driverless cars to see, lidar is used to create fast and accurate 3D scans of landscapes, buildings, cultural heritage sites and foliage. It is also used to help create Radiohead’s House of Cards music video. When positioned on top of a vehicle, it can scan up to 60 metres in all directions, generating precise 3D maps of the car’s surroundings and ensuring the vehicle can avoid obstacles and collisions. It is expensive, but it provides visibility where other sensors might fail. Lidar is the best of both the worlds—it sits between cameras and the radar, and can detect both distance and objects, and can make out the shape of those objects. Radars, on the other hand, are good at detecting objects and how far away these are, but do not offer any information about the shape or size of the object. The radar in Tesla Model S likely did detect the truck it collided into, but the system is designed to tune out objects that might look like overhead road signs to prevent it from events. The integration Richard Wallace of Tesla referred to is the algorithms and intelligence that control the way different sensors work together. Lidar and vehicle-to-vehicle communication, where each car communicates its location to others nearby, will both play a key role in building safer self-driving fleets. Lidar units that Google uses in its selfdriving cars cost up to US$ 70,000 per unit, though there are now units that cost as little as US$ 250. This could make it more accessible for the mass market. SENSORIS data standard will enable

June 2017 | electronics For You

driverless connected vehicles to prepare for changing conditions and hazards well before the vehicle, be it a truck or car, can see these. In addition to the various sensors such as lidar systems, cameras and ultrasonics required in driverless vehicles, the map is a virtual sensor that provides vital information to the vehicle about the road and terrain ahead and what it is likely to encounter over the hill or around the corner where the sensors cannot currently see. But driverless vehicles need more than just these sensors to provide a smooth and safe driving experience. These need to communicate with all other vehicles on the road by sending, receiving, interpreting and responding to live route conditions in real time. Sharing of data can apply to all modes of transport including bikes, buses, trams and trains; and not just cars or trucks. To get this right, there is simply insufficient information available from one car brand or model, and to enjoy the huge benefits of the new innovation and its reduction in emissions and congestion, we have to think both on a new scale and collaboratively. First of all, a lidar sensor is typically installed on self-driving vehicles that emit eye-safe laser light as the sensor head rotates. Unlike streetlight cameras, these systems do not read licence plates, and laser light scatters/reflects off the windshield and surrounding objects. Mobile active and ive sensors. This approach can place very inexpensive smart RF transceiver systems in millions of new and existing cars that communicate with a stationary light-pole-mounted lidar system. The easy-to-integrate embedded sensor exchanges RF signals and display in-dash and audible warnings to drivers, while the lidar system maps the local area. Like radar detectors, it could have a huge potential market and also allow for faster technology disruption on a wider scale. It could also help accumulate millions of hours of actual www.eFYmag.com

AUTOMOTIVE use statistics nationwide in hundreds of cities and municipalities at a much faster pace. Widespread embedded safety alert systems. This approach allows ad-hoc technology disruptors and creative individuals to embed these sensors more broadly in smaller and safer stationary use cases, creating smart networks of lidar and embedded sensors. A public safety system like this might provide a new kind of stationary collision-avoidance and weather systems for congested roads, toll ways, on-ramps/off-ramps, alerts for drivers about pedestrians, school crossings, bikes, narrowed lanes, accident avoidance, construction and wildlife. By sensing the automobile behaviour, it could alert drivers of roadway weather conditions (black ice, ice on bridges, snow conditions, slick roads, etc). All modes of traffic and weather conditions could be understood more deeply. In this case, any new technology ideas in these application areas would be better than none—the situation we have today. These sensor systems are continuing to get smaller and will be ideal for a number of embedded applications. Instruments and techniques such as the com, sextant, LORAN radiolocation and dead reckoning. These are among those that have been used with varying degrees of accuracy, consistency and availability. For autonomous vehicles, the navigation and guidance sub-system must always be active and keep checking how the vehicles are doing versus the goal. For example, if the originally optimum route has unexpected diversions, the path must be re-computed in real time to avoid going in the wrong direction. Since the vehicles are obviously constrained to roadways, this takes much more computational effort than simply drawing a straight line between A and B. The primary sub-system used for navigation and guidance is based on a GPS receiver, which computes the present position based on complex www.eFYmag.com

analysis of signals received from at least four constellations of over 60 low-orbit satellites. A GPS system can provide location accuracy of the order of one metre (actual number depends on many subtle issues), which is a good start for the vehicle. Note that, for a driver, who hopes to hop in the car and get going, a GPS receiver takes between 30 and 60 seconds to establish initial position, so the autonomous vehicle must delay its departure until this first fix is computed. GPS sub-systems are now available as sophisticated system-on-chip (SoC) ICs or multi-chip chipsets that require only power and antenna, and

Fig. 4: A complete GPS module

include an embedded, applicationspecific compute engine to perform intensive calculations. Although many of these ICs have an internal RF preamp for the 1.5GHz GPS signal, many vehicles opt to put the antenna on the roof with a co-located lownoise amplifier (LNA) or RF preamplifier, and locate the GPS circuitry in a more convenient location within the vehicle. The antenna must have righthand circular polarisation characteristics to match the polarisation of GPS signals, and it can be a ceramic-chip unit, a small wound stub design or have a different configuration. An example of a GPS module is RXM-GPS-F4-T from Linx Technologies, shown in Fig. 4. This 18mm×13mm×2.2mm surfacemount unit requires a single 1.8V supply at 33mA, and can acquire and track up to 48 satellites simultaneously—more channels allow the GPS to see and capture more data and, thus, yield better results and fewer dropelectronics For You | June 2017

65

AUTOMOTIVE

Fig. 5: MEMS devices have changed the implementation of IMU functions, such as gyroscopes and accelerometers

outs. Its sensitive front-end requires signal strength of -159.5dBm for operation. After it computes locations based on GPS received signals, it provides output data to the system processor via the serial interface using industrystandard National Marine Electronics Association message format. While GPS is an essential function for autonomous vehicles, it is not sufficient by itself. The GPS signal is blocked by canyons, tunnels, radio interferences and other factors, and these outages can last for many minutes. To supplement the GPS, the autonomous vehicle uses inertial guidance that requires no external signal of any type. The inertial measurement unit (IMU) consists of a platform fixed to the vehicle, and this platform has three gyroscopes and three accelerometers, one pair oriented each for orthogonal x, y and z axes. These sensors provide data on the rotational and linear motion of the platform, which is then used to calculate the motion and position of the vehicle, regardless of speed or any sort of signal obstruction. Note that, an IMU cannot tell you where you are, only the motion, so the initial location of the vehicle must be determined by the GPS or entered manually. The in-vehicle IMU would not be practical without the development of MEMS based gyroscopes and accelerometers. The historical and fully-refined IMU is based on 66

June 2017 | electronics For You

a spinning-wheel gyroscope and a gimbaled platform, which has served many applications quite well (missile guidance/Space missions), but it is simply too large, costly and powerhungry for an autonomous vehicle. A representative MEMS device is A3G4250D IC from ST Microelectronics, a low-power three-axis angular rate sensor that provides a high degree of stability at zero rate level and with high sensitivity over temperature and time, shown in Fig. 5. It provides 16-bit digitised sensor information to the ’s microprocessor via a standard SPI or I2C digital interface, depending on the chosen version. With its tiny size of just 4mm×4mm, operation from a 1.8V supply, and stability and accuracy specifications, it is well-suited for inertial automotive navigation when combined with a three-axis accelerometer, for a complete six-axis IMU. The autonomous car must be able to see and interpret what is in front when going forwards (and behind when in reverse, of course). It is also necessary to see what is on either side. In other words, it needs a 360-degree view. An array of video cameras is the obvious choice, plus a camera to determine where the lane is and sense objects or markers on the road. But using cameras alone presents problems. First, there are mechanical issues of setting up multiple cameras

correctly and keeping these clean. Second, heavy graphic processing is needed to make sense of images. Third, there is a need for depth perception as well as basic imaging. And finally, conditions of lighting, shadows and other factors make it very challenging to accurately decide what the camera is seeing. Instead, the primary vision unit on the autonomous vehicle is a lidar system. To enable the split-second decision-making needed for self-driving cars, a lidar system provides accurate 3D information on the surrounding environment. Using this data, the processor implements object identification, motion vector determination, collision prediction and avoidance strategies. Lidar unit is well-suited to big-picture imaging, and provides the needed 360-degree view by using a rotating, scanning mirror assembly on the top of the car. Lidar provides raw information using high-speed, high-power pulses of laser light that are timed with the response of a detector to calculate the distance to an object from the reflected light. An array of detectors, or a timed camera, can be used to increase the resolution of the 3D information. The pulse is very short to enhance depth resolution, and the resulting light reflections are used to create a 3D point-like cloud that is analysed to transform the data into volume identification and vector information. The transformed result is then used to calculate the vehicle’s position, speed and direction, relative to these external objects, to determine the probability of collision and instruct appropriate action, if needed. For close-in control, such as when parking, lane-changing or in bumperto-bumper traffic, lidar system is not as effective. Therefore it is supplemented by radars built into the front and rear bumpers, and sides of the vehicle. Operating frequency for this radar is usually 77GHz, which has been allocated for this use, has good RF propagation characteristics and www.eFYmag.com

AUTOMOTIVE

Fig. 6 (a): Radar system overview of AD8283, a six-channel radar receive path AFE

Fig. 6 (b): Simplified block diagram of a single channel of AD8283

provides sufficient resolution. To fit the radar into the flat bumper assembly and its limited space, it is necessary to use a highly-integrated design, including using part of the radar sub-system PC board as its antenna. Also required are active components such as AD8283 from Analog Devices, which integrates six channels of LNA, a programmable gain amplifier (PGA) and an ant ialiasing filter (AAF) plus one directto-ADC channel, with a single 12-bit analogue-to-digital converter [(Figs 6 (a) and 6(b)]. Automotive radar systems require sophisticated, controllable analogue front-end circuitry to handle reflected pulse signals across multiple receiver channels, and AD8283 from Analog Devices, which is designed specifiwww.eFYmag.com

cally for this situation. Primary application for AD8283 is in high-speed ramp, frequencymodulated, continuous-wave radar (HSR-FMCW radar). Performance of each functional block is optimised to meet the demands of this radar system with a careful balance among parameters such as LNA noise, PGA gain range, AAF cutoff characteristics, and ADC sample rate and resolution. The AD8283 includes a multiplexer in front of the ADC, which automatically switches between each active channel after each ADC sample has been taken. Each channel features a gain range of 16 to 34dB in 6dB increments and an ADC with a conversion rate of up to 72MSPS (mega samples per second). electronics For You | June 2017

67

AUTOMOTIVE isolated sensing, for safety and functionally, but isolation is not needed on lowvoltage circuit boards. Instead, the most common technique used to determine current at a source or load is with a high-side, currentsense, milli-ohm resistor (called a shunt) in Fig. 7: Simplified schematic of INA250 from Texas Instruments. conjunction with a difMeasuring current and, thus, power is a vital housekeeping ferential amplifier that function needed in most circuits measures the voltage drop across it. Although the amplifier is used with a discrete sense resistor, there is now an alternative that saves space, minimises errors in readings that are primarily due to the thermal drift of the sense resistor as it self-heats and simplifies the bill of materiFig. 8: A distributed solution als by reducing the number of parts. INA250 from Texas The combined input-referred noise Instruments puts a sense resistor and voltage of the entire channel at differential amplifier in a single packmaximum gain is 3.5nV/√Hz, which age, resulting in a far-smaller boardis a critical threshold parameter for layout footprint, fewer circuit-layout effective performance. problems and lower system cost due While components and subto simplified schematic (Fig. 7). systems used for navigation and The INA250 current-sense resistor guidance, or for image-capture and plus differential amplifier composensing, get the most attention due nents ease the design and PC-board to their glamour aspects, a large layout tradeoffs while guaranteeing portion of the design of an autonohigh precision and accuracy along mous vehicle involves mundane with lower cost. issues such as power management. Several application-specific, unique A distributed solution circuit boards and sub-systems are added to a conventional vehicle to The complex distributed solution provide the functions needed for included the architecture of a Web autonomous operation. Much of based tool for manual and semi-autothe system-level operation inmated ground-truth data validation. volves measuring and managing This Web based tool included a server the power requirements to control application, a pre-processing unit and power, overall consumption and a graphical interface that would thermal dissipation. provide the main interface for human Monitoring the current and volttesters. In addition, Saguaro provided age at the batteries often requires data inspection for, and delivery of, 68

June 2017 | electronics For You

www.eFYmag.com

AUTOMOTIVE validated ground-truth data. To achieve the results, recordings were broken into individual images and then software tools were applied to present and edit the images very rapidly and accurately. Some technologies and techniques used for this solution were: 1. Distributed software applications 2. Distributed databases 3. HADOOP clusters 4. State-of-the-art JavaScript frameworks for the interface, using three different backend approaches: based on Windows and using C#, based on Linux and Java, and based on Linux and Python The complex distributed solution comprised three main components: a server backend, database and Web client application, all designed to allow for interaction with human testers. The server backend is responsible for connecting and getting input data from company servers. The server application streams data from a company server, separates streamed data into smaller pieces using the company’s API, assigns frames to individual clients, serves more than 100 client applications with requested data, aggregates output results from Web clients and sends the results to company servers. Client applications can be configured to send data in pre-defined sequences such as timecontinuous or random. The database is used by the server backend to store multiple types of data, including Big Data from numerous environmental sensors. The Web client application enables the human tester to validate the target features identified in the current frame. Putting together this solution requires engineers to use and coordinate multiple technologies including: 1. C#/.Net 4.5 2. Entity Framework 3. Log4Net 4. XML 5. MD5 encryption for DB credential 6. Windows Server 2012 R2 www.eFYmag.com

7. HTML5 8. CSS 9. JavaScript 10. Model/View/View Model→JSON/ React (Bootstrap)/Backbone 11. NodeJS Although this particular solution is designed to meet the specific needs of advanced driver assistance systems and ground-truth validation, similar technologies, methodologies and technical capabilities can be used to develop other complex sensor based and machine learning applications. The most common applications could include solutions for the IoT. Car manufacturers have made significant advances in the past decade towards making self-driving cars a reality. However, there still remain a number of technological barriers that manufacturers must overcome before self-driving vehicles are safe enough for road use. GPS can be unreliable, computer vision systems have limitations to understanding road scenes and variable weather conditions can adversely affect the ability of onboard processors to adequately identify or track moving objects. Self-driving vehicles are also yet to demonstrate the same capabilities as human drivers in understanding and navigating unstructured environments such as construction zones and accident areas. These barriers, though, are not insurmountable. The amount of road and traffic data available to these vehicles is increasing, newer range sensors are capturing more data and algorithms for interpreting road scenes are evolving. Transition from human-operated vehicles to fully self-driving cars will be gradual, with vehicles at first performing only a sub-set of driving tasks such as parking and driving in stop-andgo traffic autonomously. As the technology improves, more driving tasks can be reliably outsourced to the vehicle. electronics For You | June 2017

69

data interface

HSIC Versus USB Andrew Rogers is applications engineer at Microchip Technology

H

igh-speed inter-chip (HSIC) interface is becoming more popular due to its notable advantages over USB for hard-wired inter-chip applications. The interface is a two-signal, source-synchronous interface that can provide USB highspeed data at 480Mbps. Data transfers are 100 per cent host-driver compatible with traditional USB topologies. Full-speed (FS) and low-speed (LS) are not ed by the format. However, a hub with HSIC can provide FS and LS . The interface differs from USB in the physical layer only. Significant features include no chirp protocol, source-synchronous serial data transmission and no hot removal or attach as the interface is always connected. It has 1.2V signal levels designed for low-power applications at standard LV

CMOS levels. Maximum trace length is 10cm. The protocol for data transactions between host and device via HSIC is the same as USB, as shown in Fig. 1. The primary difference between the two is that in HSIC all information is transmitted via a single data line, and a strobe signal communicates when to sample the received data signal. HSIC uses double data rate (DDR) signalling; data are sampled at both the rising and falling edges of the strobe signal. The strobe signal oscillates at a frequency of 240MHz, which provides a total data rate of 480Mbps.

Advantages of HSIC over USB HSIC has significant advantages over USB. For a start, it is a fully-digital standard and, thus, no analogue frontend is required. Lack of an analogue frontend means die sizes can be reduced and, thus, so can the cost. Additional die reduction can also be made due to the decreased amount of digital logic required by the simplified connection protocol. HSIC standard does not inherently reduce power consumption, but removal of the analogue frontend can lead to lowerpower designs, especially since analogue circuitry does not necessarily scale one-toone with digital circuits for reductions in process feature size. HSIC is especially low-power when placed into the suspended state as there is no current drawn on the strobe or data lines. By comparison, standard USB draws a minimum of 200μA on D+ through a 1.5kΩ pull-up resistor when suspended. Because HSIC is only different from USB at the physical layer, migrating from USB to HSIC is not like changing to a completely new standard. This means existing USB software stacks and USB protocol knowledge bases can be quickly transitioned to HSIC.

Data sampling Fig. 1: Data packet transferred from host to device

72

June 2017 | electronics For You

With standard USB, every data packet begins with a sync pattern to allow www.eFYmag.com

data interface the receiver clock to synchronise with the phase of incoming data. The differential sign of D+/D- signal is then sampled according to the sync pattern. HSIC uses a separate strobe line to tell the receiver when to sample incoming data. HSIC data signal is sampled at the rising and falling edges of the strobe signal. If the strobe and data signals become skewed for any reason, sampled Fig. 2: Equal HSIC trace lengths data may become corrupted. HSIC Electrical Specification defines the maximum allowable skew as 15ps. To make sure the skew does not become an issue, HSIC traces must be kept as short as possible and must not be longer than 10cm. Data and strobe traces must be the same length, and these should be routed to 50Ω singleended impedance. To illustrate the amount of skew possible Fig. 3: Strobe trace 10cm longer than data trace in the real world, Fig. 2 deciphered by placing a differential shows the beginning of a test packet probe connected to an oscilloscope transmitted from a host to a device at either the transmitter side or the with equal lengths. receiver side. HSIC signals are more The same packet transmitted sensitive and, thus, transmission from the same host with a strobe line theory should be considered trace that is about 10cm longer than when attempting to probe these. the data trace is shown in Fig. 3. A good general guideline is to The resulting skew is about half of probe at the side opposite to the a nanosecond. This is an extreme source of the signal that needs example, but results suggest that to be observed. For instance, to even a small amount of length observe the signals originating from mismatch may result in an HSIC a device, place a probe at host-side specification violation. terminals. To observe the signals The single-ended nature and diforiginating from a host, place the ferences in signal termination cause probes at device-side terminals. some difficulties when attempting When attempting to probe to probe HSIC lines. Standard USB signals originating from a device signals can be easily monitored and www.eFYmag.com

while probing at the device side, the signal becomes distorted. This is likely due to the interference caused by the signal reflecting back on itself. The middle of the trace can also be probed, but results are typically not as clean as if probed properly from one side. The ideal would be to probe simultaneously from both ends. A series protocol analyser may be able to sample the signals accurately in both directions, but the 10cm trace length restriction makes this option impractical.

Making the connection HSIC interface is structured such that a host or peripheral can be powered on in any order. To ensure a false connection is not detected, the host, hub and peripherals must ensure that the strobe or data lines do not float to an undetermined value, commonly referred to as tri-stated. Fig. 4 shows an oscilloscope capture of a connect sequence. This connect sequence is much simpler than the USB connect sequence because there are no speeds to negotiate. This sequence can be handled by a very simple state machine, reducing die size requirements. With standard USB, the host can determine if a downstream port has been disconnected by monitoring the magnitudes of DP/DM signal voltages. If the voltage exceeds the disconnect voltage threshold, the host can conclude that the device has been disconnected. electronics For You | June 2017

73

data interface

Fig. 4: Connect sequence from idle and suspend to connecting and resuming signalling

HSIC does not a disconnect protocol because it is intended to be a hard-wired, always-on connection. However, it is still possible to have a situation where a downstream device may appear to have disconnected, and care must be taken to ensure the host does not permanently lose its connection with the device. This apparent disconnection or standoff can occur because the host always maintains an idle state while the bus is unused, and the idle state is identical to the suspend state from a signal perspective. The host has no way of knowing if or when a downstream device has been powered down or disconnected. Since the suspend signalling is identical to the idle signalling, it is possible to reach a state where a downstream device believes it has been suspended while the upstream host thinks there is no device downstream and waits indefinitely for a connect signal to arrive. A similar standoff condition could occur if the upstream host disables the port while the device believes it has been suspended. This condition is not likely to occur between hosts and devices that 74

June 2017 | electronics For You

never cycle power or soft reset. If this issue is encountered, it must be dealt with in an application-specific manner at either the link or software-stack level. This can be done by programming the software stack or deg the link in a way that prevents the condition from occurring in the first place. Alternatively, the system(s)on-chip (SoC) can attempt to deal with a downstream device after it disconnects by resetting HSIC hub. The device discovery sequence will occur, and the connection will be re-established. On Microchip’s USB254x, USB3613, USB3813, USB4604 and USB4624 devices, the SoC can use VBUS_- DET pin to re-establish the connection. Pulling the pin low suspends the hub, while pulling the pin high wakes it up.

Conclusion HSIC standard has advantages over USB in hard-wired applications as long as the correct connection and disconnection procedures are observed. These procedures are particularly important when troubleshooting some issues involving HSIC connectivity. www.eFYmag.com

software

Open SOurce SOftware You Can Use For Iot DevelOpment

T Rajesh Sola is a faculty member of C-DAC’s Advanced Computing Training School, and an evangelist in the embedded systems and IoT domains

www.efymag.com

he Internet of Things (IoT) is spread across various layers, with different types of software being used across these. This article presents a list of open source software packages that can be used in each of the layers. You might have read Gartner and many other business forecasts on connected devices and the Internet of Things (IoT) growth in the coming years. Billions of connected devices around us, as well as open hardware and software solutions, would play a key role in enabling a smarter life in the near future. Before proceeding to the list of open source software in this field, let us take a look at a few layers of the IoT architecture. Layer 1 consists of end devices talking to the environment, which are also known as sensor nodes, typically powered by microcontroller based targets, wearables and, sometimes, Linux boards. Layer 2 consists of gateway devices, typically Linux-powered target boards or handheld gadgets, for interfacing end devices with Cloud platforms and bridging protocols with their middleware . High-end microcontrollers with suitable frameworks can also be a choice here. Layer 3 consists of servers, typically deployed on Cloud platforms. These provide services like authentication, data collection, data processing, analysis, storage, visualisation, Web integration, mobile app connec-

tivity and more. This may be further divided into various sub-layers or components. Let us examine some open source choices for development in each layer.

Layer 1 Toolchains. GNU ARM embedded toolchain provides pre-built toolchains as well as source codes for Cortex-M and Cortex-R processors, especially for bare metal code for microcontrollers. It has consistent releases in every quarter. The latest stable release is 2017 Q1, which ships with gcc v6.3.1 and is migrated to 64-bit Linux hosts, whereas the previous release of 2016 Q3 ships with gcc 5.4 and is available for 32-bit Linux. The GNU ARM Eclipse project provides Eclipse CDT extensions and for popular targets like STM32 F series boards, and FRDM Kinetis KL series boards with C/C++ templates, using this toolchain as the backend. Most of these elements are licensed under GPL. URL: developer.arm.com/open-source/ gnu-toolchain/gnu-rm, gnuarmeclipse. github.io mbed OS. This is specifically designed for IoT needs with common C++ APIs for Cortex-M architecture from various families, and is licensed under Apache 2.0. It comes with an online IDE as well as offline components for development with for various toolchains, IDEs (like Keil uVision, Eclipse, ARM GCC and IAR) and rich libraries. It accelerates IoT development with good connectivity solutions for Ethernet, Wi-Fi, Bluetooth LE, Thread, 6LowPAN, LoRaWAN, Cellular, NFC and so on. URL: developer.mbed.org Zephyr OS. This Linux Foundation Project is aimed at being a small, scalable, real-time operating system (OS) for connected and resource-constrained devices, licensed under Apache 2.0. The recent release of version 1.7.0 comes with a unified kernel, discarding the dual kernel electronics for you | June 2017

75

software (micro, nano) approach from previous releases. It can be a perfect choice for developers who are fond of Linux with its programming model for drivers, device interfaces and highly-configurable services, all in a single address space. URL: www.zephyrproject.org Platform.io. This next-generation IDE, with an ecosystem for IoT development, is based on the popular Atom editor with a cross-platform build system and library manager, licensed under Apache 2.0. It wraps popular frameworks like CMSIS, Arduino, mbed, Energia, ST Standard Peripheral Library and WiringPi, and s native applications on Linux and Windows. URL: platformio.org Arduino IDE, forks and add-ons. This open source physical computing platform has a simple IDE and coding style. Its power is rapidly getting enhanced by available add-ons for other families of boards, like ESP8266, ESP32 and NRF5x series. TI Enregia, a fork of Arduino, is available for popular TI launchpads like CC3200, MSP series and Tiva series, with rich for board peripherals and IoT connectivity. Most of these elements are licensed under GPL 2. URL: arduino.cc, energia.nu

Layer 2 NodeRED. This is a visual tool for wiring hardware peripherals and online services, licensed under Apache 2.0. It has a rich collection of nodes for sensor interfacing, local connectivity (serial, Wi-Fi, Bluetooth, etc), Cloud connectivity (HTTP, mqtt, etc) and social media services. It can run on any OS with Node.js runtime or in a Docker container. The latest Raspbian for Raspberry Pi and Debian for BeagleBone Black ships with NodeRED, by default. It is also available from Cloud-hosted instances like IBM Bluemix and Sensetecnic FRED. It can even communicate to Arduino-like targets using 76

June 2017 | electronics for you

Firmata protocol. Custom nodes can be built with ease using Node. js backend and HTML frontend. It is a perfect choice for kickstarting IoT prototyping with zero or little programming effort. URL: nodered.org, flows.nodered. org Eclipse Kura. This is an OSGi based container providing various gateway services addressing M2M and IoT needs, licensed under Eclipse public licence 1.0. Kura components are highly configurable through a Web console, which can be done dynamically. Kura s good connectivity like CANbus, Serial Bus, Bluetooth LE, Cloud services via MQTT, etc, and even for hardware access using OpenJDK Device I/O library. It aims at reducing the barrier between enterprise and embedded systems with a rich set of Java APIs. URL: eclipse.org/kura/ Macchina.io. This is a toolkit for building embedded applications for IoT on top of POCO C++ libraries and V8 JavaScript engine, licensed under Apache 2.0. The core is implemented in C++ for maximum efficiency and JavaScript is used for application development with its ease of programming. The foundation for macchina.io is Open Service Platform (OSP), which enables dynamically extensible, modular applications based on the powerful plug-in and services model similar to OSGi in Java. URL: macchina.io Eclipse SmartHome and OpenHAB. Eclipse SmartHome is an OSGi based framework for smart home solutions. It provides a rich set of OSGi bundles for various services and a high degree of modularity. OpenHAB is a usable product design based on SmartHome framework. It is vendor neutral and hardware/technology agnostic with automation software for rapid development of smart home solutions. Both are licensed under Eclipse public licence 1.0.

URL: eclipse.org/smarthome, openhab.org Iotivity. This Linux Foundation project is initiated by Open Connectivity Foundation (OCF), formerly known as Open Interconenct Consortium (OIC), which comprises a group of companies including Samsung and Intel. OCF aims at seamless device connectivity and standardisation of communications for billions of devices. A reference implementation of this standard is released with features like powerful device management, resource management, services and security, as well as communication based on CoAP, and has been released under Apache 2.0 licence. This project also provides a few modules for building cloud services for interconnecting Iotivity clients. URL: iotivity.org Contiki OS and RIOT OS. These operating systems are designed for devices with constrained network and memory resources. These come with full IPv4- and IPv6-capable stacks and low-power wireless standards such as 6LowPAN and RPL, and protocol connectivity like CoAP. Contiki is licensed under BSD 3-Clause and RIOT under GPL 2.1. Cooja simulator provided by Contiki allows testing of application developments for large wireless networks with emulation of various hardware targets. RIOT is a -friendly OS similar to Contiki and comes with additional offerings like a minimal footprint, C++ APIs, threading and real-time capabilities with minimal low overheads. URL: contiki-os.org, riot-os.org Note. Even though some level of classification is done based on design goals and primary usage, certain elements are used across layers. For example, Contiki and RIOT can also be used in Layer1, and mbed.org and Zephyr OS can be used in Layer 2, and so on.

Layer 3 Kaa IoT. Kaa is a multipurpose middleware platform for connected www.efymag.com

software things and enabling end-to-end IoT solutions. It can be deployed on Amazon Web Services or your own server. A rich set of SDKs can be generated from a Kaa server with APIs in different languages for seamless device connectivity. It can integrate with many databases, data processing elements like Mongodb, Apache Spark and Cassandra with for structured and unstructured data and messaging frameworks like Apache Kafka. It enables highly-available servers with clustering , thus reducing downtime and precluding single points of failure using Zookeeper . URL: kaaproject.org SiteWhere. This is an open platform for the IoT with good protocol and powerful device and asset management services, and is licensed under common public attribution licence 1.0 (AL-1.0). It is designed as a multi-tenant system to host multiple applications in a single instance of deployment. Popular open source solutions like Apache Spark, OpenHAB and InfluDb/Grafana can be integrated by SiteWhere. It can be deployed on public Clouds such as Azure and Amazon EC2, service providers based on Ubuntu Juju and Docker, or your own server. URL: sitewhere.org InfluxData TICK stack. InfluxData provides TICK stack with four powerful components for time series data management. It is written in Go language and licensed under MIT . The components are: 1. Telegraf, used for data collection and publishing metrics with a rich set of input, output and service plugins. 2. InfluxDb, a time series database with high availability and high performance without any external dependencies (unlike other open source time series models like OpenTSDB based on Hbase or KairosDB based on Cassandra). It has good for down-sampling high-precision data and data-retenwww.efymag.com

tion policies. 3. Chronograf, a visualisation tool for exploring data. 4. Kapacitor, which is used for data processing with alert management and ETL jobs. URL: influxdata.com Mosquitto. This is an Eclipse IoT project implementing MQTT broker with version 3.1.1 and 3.1 for backward compatibility. It is licensed under Eclipse public licence 1.0 and EDL 1.0. It also provides client libraries and powerful reference clients. Popular Cloud services like CloudMQTT are built on top of Mosquitto. URL: mosquitto.org CloudFoundry. This open source PaaS platform is a Linux Foundation collaborative project, and has been released under Apache 2.0 licence. It is available as a service from a few commercial providers like Bluemix or as an open source product to be deployed on one’s own servers. It s the full life cycle of application development in various languages for IoT needs, and comes with good service like data storage, messaging, application development and APIs for mobile apps. URL: cloudfoundry.org

Connectivity Paho. This is an Eclipse IoT project offering MQTT clients, licensed under EPL 1.0 and EDL 1.0. It offers synchronous and asynchronous API modes in various languages like C, C++, Java, Python, Go, JavaScript and .Net (C#). It also provides Android service for building mobile apps and reference apps, which can be customised according to an application’s needs. URL: eclipse.org/paho Eclipse Californium. This is divided into a few sub-projects. The core comes with Java API model for CoAP client and server design, and a few reference apps. It can be built as a Maven project and can be embedded in other Java applications.

The Scandium sub-project implements DTLS 1.2 for secure CoAP. It also provides a proxy library and reference app for CoAP-HTTP bridging. These are dual licensed under EPL 1.0 and EDL 1.0. URL: eclipse.org/californium Bluez and bindings. This is an official Linux Bluetooth protocol stack that comes with a space library and tools. It s various classical protocols like L2CAP, RFCOMM and SDP, as well as LE protocols and profiles like GATT and GAP. It has Python bindings via PyBluez and Node.js packages. Bleno and noble via npm are available for BLE advertising in peripheral mode and discovery in central mode. These node packages are built on top of Bluez. URL: bluez.org, karulis.github.io/ pybluez/, github.com/sandeepmistry CETIC 6LBR and Linux WPAN. 6LowPAN/RPL Border Router (6LBR) can interconnect end devices running on 6LowPAN with the Internet by bridging 802.15.4 network with IPv6 on the gateway side. It s various 15.4-capable targets like OpenMote, TI CC25xx/26xx platforms and Linux hosts like Raspberry Pi, and can run in Bridge or Router modes. It is based on Contiki project and licensed under similar . Linux WPAN enables 802.15.4 stack at the kernel level and space tools for 6LowPAN based development on Linux. URL: http://cetic.github.io/6lbr/, github.com/linux-wpan Wireshark filters. Wireshark, a well-known network protocol analyser and packet generator, is licensed under GNU GPL. It has rich filter s for Internet protocols like MQTT, CoAP, HTTP and Websockets. It can also analyse Bluetooth (classic, LE) packets and IEEE 802.15.4 traffic. This is a reprint of the article published in February issue of Open Source For You magazine. electronics for you | June 2017

77

led lighting

LED FiLamEnt BuLBs Growing Your Business with Aesthetics

A Vertica Asthana is a technical journalist at EFY

jeweller recently replaced the traditional incandescent lights of his store with LED filament bulbs. Ask him why he did it and pat comes the reply: “I wanted my diamonds to shine bright and look attractive to the customers.” LED filament bulbs are taking the market by storm. Enterprises in hospitality, lifestyle and jewellery sectors have been increasingly replacing their Edison bulbs (the regular incandescent bulbs that were invented by Thomas Edison) with this new kind to enhance their ambience and get more customers.

What makes LED filament technology special The LED filament technology is quite simple, using a chip-on-glass and a particular mixture of gases inside the glass bulb. The chip-on-glass technology uses a

78

June 2017 | electronics For You

glass substrate instead of the traditional metal substrate to mount unpackaged LEDs, which helps the LED filament bulbs to emit light uniformly without losses. Once LED chip bonding is done, it is encapsulated in a phosphor material, which affects the colour of the light emitted by the bulb. Some manufacturers also use this phosphor for temperature control of the product. Vinod Kumar, vice-president, R&D, at Surya Roshni, shares that the circuit and driver technology has not changed so far. “Design engineers had to work really hard with the technology to make it an electromagnetic-compatible product in the same form factor, which results in lower fluctuations,” he adds. The proportion of gases inside the glass bulb decides the temperature of the product. This, in turn, directly affects the longevity of the bulbs. LED majors, however, rely more on the high-voltage,

www.eFYmag.com

led lighting low-current combination technique—a major reason why temperature control is possible in LED filament bulbs.

The current scenario While the market for LED filament bulbs is still developing in India, these bulbs have a growing demand in overseas markets such as Europe. LED filament bulbs come in various shapes (oval, globe and candelabra) and sizes. At present, these are available in warm white and cool white colours.

Making the switch: From Edison to LED LED filament bulbs offer a variety of benefits to the customers, making them a good choice for replacement. Cost-efficiency is their USP. Eliminate frequent bulb replacement. One of the many concerns in running a business is keeping the operational costs low. And this includes the cost of the lighting system installed in the store. The heat issue and power output of light have always affected air-conditioning and performance of the bulbs used. In an incandescent bulb, electricity flowing through the filament generates heat, which turns the filament red. The entire process results in decreasing the life-hours of the bulb. This, in turn, forces the retailer to replace the bulbs more frequently, thus incurring unnecessary costs. However, LED filament bulbs don’t heat up much, which ensures their longevity. In of life-hours, an LED filament bulb can light up for a whopping 25,000 hours, while an incandescent bulb lights up for 1000 hours only. Make it brighter. In an incandescent bulb, only one per cent of the total energy dissipated is converted into light energy, thus resulting in a lower output. In contrast, LED filament bulbs give a higher lumens output as nearly 40 per cent of the energy dissipated by these bulbs is converted into light while the loss of energy is very negligible. www.eFYmag.com

For instance, if a restaurant uses traditional incandescent bulbs in its dining area, at least three to four such bulbs will be needed in just one part of the room in order to keep it bright for customers. However, when using LED filament bulbs, the restaurateur may have to fix only a single bulb in that area as the efficacy of LED filament bulbs is substantially higher than traditional Edison bulbs. Create the right ambience. Ambience plays a pivotal role in customer service oriented businesses like hospitality and retail stores. Dimmers are crucial in such scenarios. Usually, lifestyle chains, hotels and restaurants have dimmers installed in their stores to create the right ambience for their customers. But since traditional incandescent bulbs are often not compatible with these dimmers, they end up fluctuating frequently, thus rendering dimmers completely useless. Nevertheless, lighting majors have addressed this issue in LED filament bulbs, making them compatible with most of the dimmers. To understand, consider the situation in which a customer wants to have a romantic dinner with his partner at your restaurant but is put off by the 100-watt traditional incandescent lights everywhere. While you would want to serve this customer, unfortunately, you do not have the right tools to fulfil this need. Ultimately, you end up losing a customer because of your 100-watt Edison bulbs. Now, imagine dimming the lights with LED filament bulbs to create the right ambience. You have not only won a new customer but also probably a returning customer and a great word-of-mouth promotion for your restaurant. Longer durability. We explained above how LED filament bulbs have a lower maintenance cost than typical incandescent bulbs. Another contributing factor is that LED filament bulbs are far more durable than their incandescent

counterparts: The tungsten filament of incandescent bulbs is very fragile and can break even with very little mishandling of the bulb.

The glitch: high initial cost While LED filament bulbs use a high-end technology and offer several benefits to customers, unfortunately, these are exorbitantly priced and can burn a hole in buyers’ pockets. On an average, an LED filament bulb costs nearly 100 per cent more than an incandescent bulb and at least 20 per cent more than the ordinary LED bulb. Justifying the high price of the product, Kumar explains that since most lighting players do not have the requisite infrastructure to manufacture the components required for LED filament bulbs, they have to import the components from abroad. “Initially this product is going to be costly as we have signed agreements with LED suppliers and took permission for patents regarding the gas mixture to be used in glass bulb. Later, the price of the product will come down as the need for heat sinks for heat dissipation has been eliminated completely, resulting in a lightweight lamp that costs less,” he shares. At present, LED light manufacturers source LED filaments mainly from China.

Are LED filament bulbs here to stay? Industry experts suggest that the market for LED filament bulbs will grow rapidly as customers are increasingly looking at long-term benefits rather than just the initial cost. According to a market report by Epistar, the LED filament bulb market is expected to reach up to 300 million bulbs by 2017 and double to 600 million bulbs by 2018. And if the numbers are too hard to believe, wait and watch your competitors lap up rave reviews for their sudden aestheticism while you struggle to get a few more customers! electronics For You | June 2017

79

efy

dvd

Free Tools To Make Your PCB Smarter Vertica asthana

Simutron Whether you want to run your Arduino firmware or debug it, Simutron provides an integrated development environment to meet your needs. This electronic circuit simulation freeware lets you design as well as simulate circuits. Internally, it uses a Simavr AVR processor simulator. You can use Simutron for 16MHz microcontroller units. In particular, it can be used in conjunction with its CAM front-end to develop CNC firmware without access to the real hardware.

SimulIDE SimulIDE is a real-time circuit simulator freeware that s microcontrollers such as PIC, AVR and even Arduino. Whether it is a small circuit or a circuit consisting of several nodes, SimulIDE makes design and simulation quick and easy for s. It comes with a limited set of features as it is intended for hobbyists and students who simply want to experiment with basic electronics projects.

Circuit Diagram Quite useful for starters and students, this easy-to-use open source program lets s make electronic circuit diagrams and export them as images in SVG or PNG formats. They can

82

June 2017 | electronics For You

Some popular resources Speccy. Using this software you can get details of every component in your computer, from the operating system of your PC to audio , peripherals or network. The software supplements this information with temperatures of the components, so you can easily see if there’s a problem! FileZilla. It is a fast and reliable crossplatform FTP, FTPS and SFTP client. It has an intuitive graphical interface with several important features. IP Scanner. This open source freeware helps the master computer to access other PCs in the network. One just needs a free R viewer to operate a remote machine. Thus, all the systems in the network can be scanned for viruses using a friendly interface system. The can also remotely turn systems off. This software is designed for all levels of expertise. Jaspersoft. You can create web-based reports using Jaspersoft tool. It picks information from various data sources and presents it in an easy-to-read, highly interactive format for business s.

design diagrams visually by placing components with the cursor. Component values can be modified using the properties editor.

TransistorAmp Designers face issues in calculating transistor parameters like gain, amplification and biasing. What if they have a software that lets them calculate the perfor-

mance of transistors and design perfect schematics of commonbase, common-emitter and common-collector amplifiers using combinations of transistors? This freeware tool does exactly that. It shows the values of other components required to get the desired results from a given input data.

Gputils Microchip’s PIC microcontroller is used in various development boards. Gputils is a collection of tools used by PIC microcontroller for serial programming and reprogramming in flash memory. Moreover, it is compatible with Microchip’s tools MPASM, MPLINK and MPLIB. It runs on Linux, Windows as well as Mac OS platforms.

XYCE It is an open source, SPICEcompatible analogue circuit simulator, capable of solving extremely large circuit problems by ing large-scale parallel computing platforms. It also s serial execution on all common desktop platforms, and small-scale parallel runs on Unix-like systems. It uses a differential-algebraic equation formulation. Being SPICEcompatible, it s standard analysis methods. Vertica Asthana is a technical journalist at EFY

www.eFYmag.com

efy

dvd

‘Balance The Books’ With FreeCAD Vertica asthana

D

esigners struggle to find the right mechanical and electronic design and simulation software. Innumerable meetings and brainstorming over alternatives eat into their time and energy. Worse, they end up buying the software, paying a hefty amount for it. Thankfully, a saviour has emerged. FreeCAD is an open source mechanical CAD freeware that is as good as its paid counterparts. Its latest version, VER 0.16, is out in the market and ready to ease the work of designers.

A no-cost, easy-to-use tool FreeCAD is a -friendly tool with features of design, simulation and strength calculations for the product. s need no prior training to be able to use it. After design, a mesh

frame is created, which is followed by simulation (Fig. 1). Consider the case when a designer is making metallic housing for an LED street-light. The designer needs to understand areas that would be under more mechanical stress or require more strength. He also needs to understand thermal management in the housing. Also, there will be calculations related to mechanical functions. The FreeCAD software comes handy here: It would first work on the design workbench, then mesh for strengthening the design, and finally simulation.

spreadsheet format, just like in CATIA. Called Expression , this helps designers to get the precise location of coordinates. FreeCAD works on the concept of workbench. Sketcher Solver upgrade. Sketcher Solver was introduced in the previous version (0.15), but in the new version (0.16) it makes sketch editing

How it helps The latest version of FreeCAD, 0.16, has some additional features: Expression . FreeCAD helps make models using Excel

Fig. 2: Sketcher Solver (Image courtesy: www.opennet.ru)

Fig. 1: 3D view in FreeCAD (Image courtesy: www.freecadweb.org) www.efymag.com

electronics for you Plus | June 2017

1i

efy

dvd

Fig. 3: Aluminium capacitor and its mechanical view (Image courtesy: raw.githubcontent.com)

even faster and more stable than the previous version (Fig. 2). Touchscreen 3D navigation. This makes it easy to navigate. The designer can move the design in any direction with a single finger drag. So FreeCAD can be used without a mouse on a convertible laptop with touchscreen and pen, away from a desk. New tools to reduce work stress. Mechanical parts can have a regular or a discrete shape. Older versions of FreeCAD software did not take various shapes into consideration, making it a constraint for designers to work freely on any dimension.

New tools for connect, embed and cutout have been introduced in the latest version. Tools for shapes such as round, trunc, ceil and floor have also been included; version 0.15 featured shapes like parabola and hyperbola. Besides, new features have been introduced for toggle mode, continuous creation mode, speed up, duplication and mirroring (Fig. 3). Rectangles, wires and lines can now be subdivided, allowing all kinds of new shape combinations. A new spreadsheet view tool allows s to place a range of cells from a spreadsheet on a Drawing page (Fig. 4).

THE COMPLETE MAGAZINE ON OPEN SOURCE

Edit in real time. You can see the 3D view in real time. This helps you visualise the actual dimension values of a shape. The new Finite Element Method (FEM) workbench allows cleanups, while Path workbench adds editing and import. So you can change the values in real time too. Edits made are saved immediately, helping you to conceptualise in a better way.

FEM workbench GUI FreeCAD workbench consists of pre-processing, solving and postprocessing. Initially, geometry is modelled,

Your favourite Magazine on Open Source is now on the Web, too.

OpenSourceForU.com Follow us on Twitter@LinuxForYou

ii

June 2017 | electronics for you Plus

www.efymag.com

efy

dvd

he can use FEM to sort out problems like heat transfer, fluid flow or structural analysis. This is totally based on algebraic calculations. Once the designer is sure of his designed product, he can give finishing touch to it. For example, in the case of a mechanical pulley system, once the freeware finds the design perfect as per calculations, the designer can define the places where the miscellaneous parts of pulley like nuts and bolts can be tightened. Note. To read manual, please visit https://www. freecadweb.org/wiki/ Fig. 4: A new spreadsheet view tool allows s to place a range of cells from a spreadsheet on a Drawing page (Image courtesy: www.freecadweb.org) Manual#Key_features. For installation manual, refer ree files. then an FEM mesh created out of Finally, the material is added. the geometry model. Load and supThe designer prepares a frameport fixes are added to the model. work or skeleton of the design. Then Vertica Asthana is a technical journalist at EFY

www.efymag.com

electronics for you Plus | June 2017

iii

efy

dvd

Pad2Pad: Manufacturer At Your Service! Vertica asthana

D

esigners portray ideas as reality using a base. Electronic design software serves as the base, or the platform. What if a free design and simulation software gives access to customised PCBs as well? The outcome would be wonderful as designers will have a smooth sail from design to prototyping and production. That’s exactly what Pad2Pad freeware does.

The basics Pad2Pad is a computer-aided design (CAD) software that was first released in 2007. It runs on Windows 98, XP, 2000, ME, Vista Home , Vista Business and Vista Home Basic versions. You can either create in Pad2Pad, or test your designs in SWPICE program and import the net list into Pad2Pad. What is really exciting about Pad2Pad is that you can order even small quantities of boards and still get silk screen and solder masking. Also, components can be placed directly. Pad2Pad freeware is updated on a regular basis, and every new version comes with some added advantages.

How it saves time and money A well-equipped freeware. Beginners and hobbyists prefer design software that are free of extra cost and available without license, whereas professionals look out for comprehensive and well-arranged design tools. Pad2Pad is a wellstructured freeware tool that lets you access bill of materials, iV

June 2017 | electronics For You Plus

Fig. 1: Footprints and component manager (Image courtesy: www.pad2pad.com)

wherein you can add or remove components as desired. Pad2Pad s up to 16 layers. You can also build customised footprints as per component sizes. The tool s both metric and US customary units. Mechanical data export. To explain, let’s take the example of an LED light. It takes a driver PCB as well as mechanical housing, heat-sink, screws and washers to make a complete lighting module. If a design engineer does not have the facility of electronics and mechanical data export at once, there would be delay in each process—adding costs while creating clutter. The feature of

mechanical data export helps save designers’ precious time. Above all, this freeware lets you select the number of layers for your PCBs. Design and order together. After deciding on components and their pads, designers can set an electrical net. This net helps realise connections between components, reducing the chances of errors. Pad2Pad developers suggest customers to only place small orders initially. Once the test trial is successful, they can place a bigger order. This helps them avoid losses due to a defective design. During the trial lot, the customer www.eFYmag.com

efy dvd internet of things The system solution to mitiSPI Serial Flash gating something like this is to Main Storing Device Controller implement secure boot for the FW main PLC U. This is a way of authenticating the firmware and only accepting software that has a valid Powerdigital signature. Depending on MAXQ1050 Management the requirements, you could also IC encrypt the firmware. Fig. 6: Secure boot of the main PLC U Security processing demands can easily overwhelm the MIPS of a traditional PLC U or even encryption key is of prime considcreate latency issues. This is best eration in many applications, since done by off-loading the security there is no security once the key functions to a low-cost, off-theis compromised. shelf secure microprocessor that is To properly address physical built for these functions, as shown security, several issues must be in Fig. 6. The system shown here considered. These include a physical uses an external secure micropromechanism for generating random cessor to validate the firmware’s keys, a physical design that prevents digital signature. covert electronic interception of a All the above examples use key that is being communicated keys to enable authentication, between authorised agents, and a but this raises the question of key secure method of storing a key that Fig. 2: Pad2PadPhysical screen andsecurity layout (Image courtesy: www.brothersoft.com) protection. of an protects against clandestine physical

www.eFYmag.com

50

March 2017 | ElEctronics For You

can test the prototype for proper working of both the mechanical and mechanical probing. and electronic parts. Various secure key-storage devicOthersystem than regular in-built es provide designers a host features, customers need to of features that range from package specify inner layers of the stack, design to external-sensor interfaces material’s temperature strength and internal circuit architectures. and requirements other such features to get the These were developed PCB customised to their by American military in therequireform of ments. This information helps FIPS 140 standard, and many chipto avoid conflicts with customers, vendors provide very comprehensive making Pad2Pad a professional tamper-proof capabilities that can be and reliable platform. Pad2Pad used in ICSes. prototype and production services The the IoT security alsofuture followof IPC standards for safety requirements. There may be other approaches to security as well, and as you begin to An all-in-one package realise how important security is in You can design, fabricate and a connected factories environment, even with Pad2Pad. you willassemble eventually coalesce around Thisapproaches. free simulation tool also a few comes with , The IIoT incustomer manufacturing is which is a special advantage in high demand, and is a growing to designers. trend. Security will also eventually grow to cover vulnerabilities, but is a technical theVertica needAsthana is already here. journalist at EFY

electronics For You Plus | June 2017

V

www.EFYMag.coM

efy

dvd

Lock Your Design With ZenitPCB Suite Vertica asthana

T

here are a lot many PCB design and simulation software available today. What if you get a software to make schematics, layouts and Gerber files free of cost? If the software is also easy to use with an easy interface, it is definitely worth a try. Here we present such a software for electronics design, called ZenitPCB.

Why ZenitPCB ZenitPCB is a freeware comparable with other electronic design automation (EDA) software in of professional-level accuracy of designs. s can import a mechanical drawing with a PCB and build their own board outline, thereby saving time while getting a concise layout. This can be done using ‘Import DXF’ command. s find this software easy to use even without any documentation. Still, for further assistance, there is a help window F1.

Fig. 1: PCB layout designed using ZenitPCB software (Image courtesy: www.zenitpcb.com)

Version 2.0 vs 1.8 Verification and security. From copy-and-paste of terminology to modifying the selected symbols or parts, ZenitPCB is now easily accessible for commands and functions. It takes just a double click to zoom around the component that is not verified. Coming to security part, s can lock the project even when the window is open. So accidentally or deliberately, no one can change the design. Pins attribute. In pins attribute, s can insert pin names diwww.efymag.com

Fig. 2: Snapshot of ZenitPCB window

rectly inside a grid, copy and paste, eliminating the tedious process of single, manual inputs. They can also change other attributes such as ‘electrical type’ and ‘inside and outside edge.’

Auto open part or symbol. s can directly open the part required for modification. The design output form is also redesigned. Trace the current in track. ‘Current capacity’ helps to trace the electronics for you Plus | june 2017

Vii

efy

dvd

Fig. 3: Function window of ZenitPCB

Fig. 4: Auto open part

current flowing through tracks. Calculations are based on an empirical equation. Sometimes the designer is new and unaware of the flow of current at each point. It may happen that

either out of ignorance or haste, the track of a high-current path is not as per the PCB standard. If the software is not able to point out this error, the engineer would face complex electrical issues during the development stage. Deg another PCB means repeating the entire process, causing a loss of time and money loss as well. In such situations, ‘current capacity’ feature is a blessing. Export IDF and footprint wizard. Using ‘Export IDF,’ the design created can be exported into an ASCII file that contains all the 3D information. Many professionallevel CAD tools can import this file. So this feature is very helpful for mechanical designers. Customising the components is a time-taking process, demanding a high level of accuracy. The footprint wizard helps create footprints automatically as per the design requirements. Other features. Sometimes, during the nomenclature of bigger circuits, designers might repeat names on components. For example, three resistors can have a single terminology R1, which is both inaccurate and difficult to know, especially when the circuit uses a large number of components. ZenitPCB provides a feature to find such net names. Besides, there are some other small upgrades in version 2.0, which can be found in ZenitPCB suite 2.0 manual on www.zenitpcb. com/ZenitSuite_Whats_new_V20.pdf

To sum up ZenitPCB is an easy-to-use tool that does not require any special training for use. Moreover, it works on Windows platform ing 3D modelling and CAD. However, it still requires use of a mouse to design a circuit. Fig. 5: The design output form is also redesigned

Viii

june 2017 | electronics for you Plus

Vertica Asthana is a technical journalist at EFY www.efymag.com

InnovatIon

HBand: Where Your Hand Becomes The Phone

N Ankita K.S. is audience development editor at EFY, and secretary of IEEE-YP. She is an engineering graduate, and writes articles on technology for electronicsforu.com

ext time, when you see people talking with their hand on the ear with no phone but just a watch on the wrist, don’t be bewildered, they must be using the HBand. You may ask, “How can a band or watch replace a phone?” But, HBand, with its unique concept, makes it possible to talk without a phone. HBand is a luxury leather watch strap that is compatible with most of the smartwatches. It has a built-in Bluetooth audio fob, which compensates for the need of a phone during a call. This compact, ultraslim Bluetooth headset is embedded in the watch strap but can be easily removed, thereby making it both audio on the wrist and headset on the wrist.

Private calling/hands-on and headset modes

HBand in hands-on mode

84

HBand provides the convenience of phone call in two modes: hands-on mode and headset mode. Mobile phones have advanced to such an extent that call features are being provided in smartwatches. So there are no issues like the phone getting lost. Surprisingly, though, none of smartwatches provides call handling in private mode. HBand completes Apple or other smartwatches by adding the private call feature that they lack, using the patented Hands-on-Talk technology. You just have to flip open the fob and raise your hand to the ear to take the phone call. There is no need to reach for your phone and rush to pick it up. This is called hands-on mode, and is the ideal solution for a quick phone call. If you must take a longer call, you can pull out the fob from the watch strap, insert it in your ear and use the device like a normal Bluetooth headset. This is called the headset mode.

June 2017 | electronics For You

Some HBand compatible smartwatches • • • • •

Apple Watch, 42mm and 38mm models Pebble watches with 22mm band LG Urbane Moto 360-2 Classic watches or any smartwatch with 22mm or 24mm spring bar

HBand’s audio features • Hand-to-ear private call • Compact Bluetooth headset that docks on watch strap • Call and voice command • Reads out notification and time (with an app for Android) • High-fidelity music • Two-hour talk time, average two-day usage • Noise and echo cancellation • Water-resistant speaker and mic • Magnetic charger

The mechanism Directional audio is fired from the bottom of the wrist, at higher volume and pressures with sufficient back volume, towards the palm of the . When the palm is held in cupped position to the ear, it acts like a parabolic reflector, redirecting the sound to the ear. Typically, the audio is omnidirectional. Partial directivity can be given to the audio by firing at a close range with proper openings. HBand uses normal micro speakers with suitably designed back volume, front volume and openings to provide partial directivity. This method renders partial to full private sound.

Bluetooth and the microphone HBand uses the same technology as that of basic Bluetooth headsets. To achieve the higher volumes required for hands-on private calling, though, the amplification circuit of a basic headset is modified. A couple of hardware components are used www.eFYmag.com

InnovatIon to make the concept possible, the most important being a Bluetooth communication module to provide two-way audio to the phone. A directional speaker is used to fire audio to the ’s hand, while an omnidirectional microphone placed beside the speaker reads the ’s speech. Noise-cancelThe working mechanism Ram Pattikonda, the man of HBand behind HBand lation algorithms are included to cancel out the wind and other ambient noises. The watch strap uses a HBand uses a new strap clasping quick-snap magnetic buckle, which concept that accommodates the slim provides docking for the Bluetooth fob on the wrist and makes fob-open audio fob. and pull-out functions easy, without adding to the thickness on the wrist. Shrinking circuits to fit the The electronics and the audio acousmechanical complexity tics are miniaturised, while maintaining the style and multi-function “The biggest challenge was to capability. The HBand product deshrink the circuits and achieve speaker back volumes in a slim form velopment took about one-and-a-half years because of the delays in plastic factor suitable for the wrist,” shares tooling, packaging and integration Ram Pattikonda, chief technology due to the product’s mechanical officer (CTO), iMetrix Technology— complexity. The electronics system, the company behind HBand.

www.eFYmag.com

though complex, was designed in less than six months.

Essentials always on your wrist Almost all smartwatches feature music, call display, weather information and message notification. HBand enhances these features with its audio and easy-to-use mechanical structure. Its innovative combination of mechanical, electronic and acoustic features takes ordinary smartwatches to another level. The product has already been launched in India and the US. In fact, there is more to come. The next version of HBand is being designed with focus on better product design and touch-button facilities. The team is also working towards adding features like fitness and sleep tracker, high-resolution OLED display and notification with scrolling text, while extending the battery life to one week.

electronics For You | June 2017

85

innovation you can relate to

Janani Gopalakrishnan Vikram

Neil, the Little Explorer: Your child’s robot companion Livescribe smartpen instantly digitises handwritten stuff (Courtesy: Livescribe)

Livescribe: Smart pen-paper combo that saves notes digitally Even in today’s digital world, there are many who prefer a pen and paper to jot things down, whether it is a shopping list or the minutes of a meeting. The Livescribe 3 smartpen ensures that these hurriedly noted down texts do not get lost. The Bluetooth Smart enabled pen digitises notes written on a special dotted paper and saves these in your mobile device with appropriate tags, using an associated app. At first look, the smartpen appears to be a stylish, ballpoint pen with a Swiss-made tungsten-carbide ballpoint ink cartridge. Inside however is a whole lot of technology—from an infrared camera and ARM processor to a Bluetooth Smart chipset, flash memory and lithium-ion battery. An integrated stylus cap conceals the micro-USB port to charge the device. The pen works with Livescribe+ mobile app, which instantly syncs everything s write on paper to their mobile devices. The notes can be viewed in either Page View or Feed View, where they are sorted into smaller snippets. You can search, tag and convert them into text. The app also lets you add audio to the notes. Reminders, appointments and lists that you have jotted down by hand can be used to automatically set up calendar events and add new s. The content can also be converted into shareable PDF files. So that wonderful start-up idea, sketched on a paper napkin at a coffee shop, can be shared instantly with those who matter! Company: Livescribe (part of the Anoto Group AB); Country: United States of America; Website: https://www. livescribe.com/en-gb/smartpen/ls3/ 86

June 2017 | electronics For You

Working parents are often unable to spend enough time with their kids, and as a result they are unaware of their childrens’ real interests. In a simple but effective way, Neil, the Little Explorer, can help such parents. Neil is a robotic toy that keeps your child company all day long. An associated mobile app lets parents watch what their kids learn, and understand patterns about their likes and dislikes. They get to know their child’s curiosity towards a particular subject like new places, planets or biology. These patterns also help Neil to share apt information related to the child’s interests, so the interactions are more engaging. Press a button at any time and it will share a factoid (wiki as they call it) in a very conversational way. Simple button controls let Neil know if the child loved it or wants to move to the next one, revisit the previous one or get a wiki related to the current location. While the conversation is in English by default, you can buy additional language packages if you wish. Neil, the Little Explorer, chats with your kids Created by the Another Circus Deand shares factoids on various subjects sign Agency, Neil levitates a bit once in a while to remind the child of its presence. It has a pleasant LED light that lights up the child’s room at night. Charging is as simple as placing Neil on its ‘planet’ or base station. Company: Another Circus Design Agency; Country: Greece; Website: http://www. neilthelittleexplorer.com/, https://www.anothercircus.com/

ili: Wearable, instant translator Not being understood in a foreign land can be both fun and dangerous, whichever way you see it! The problem only gets worse when you travel to places where the locals do not understand English, and you are also unable to connect to the Internet to use an online translator. ili is a wearable translator that can solve this problem for you. A compact device with an in-built processor and memory, it can translate conversations instantly without requiring an Internet connection. It translates using an in-device database, and can do its job in just 0.2 seconds! ili’s dynamic dual microphone enables noise cancellation, and smart amplifiers and speaker ensure clear sound output. The sound is captured clearly, translated and instantly output in the other language.

ili is a wearable translator that lets you communicate in English, Japanese and Chinese instantly without Internet or Wi-Fi connection (Courtesy: Logbar)

Likewise, replies are also translated, so the conversation is completely understandable to both parties. Currently, ili works in English, Japanese and Chinese but there are plans to include more languages, too. What makes ili so innovative is that it is a standalone device that does not require a mobile app, Internet or Wi-Fi. It is truly a hassle-free and easy-to-use translator! Company: Logbar; Country: Japan; Website: http://logbar.jp/en/index.html, http://iamili.com/ www.eFYmag.com

InnovatIon

Kuri: A friendly home-assistant and security guard For $699, you can now own a robot that does more than just keep an eye on your house. Kuri is a 50cm tall robot that moves around your house like one of the family. With Wi-Fi and Bluetooth connectivity, it is always in touch with you. Its eyes, with built-in high-definition cameras, can be your eyes when you are away from the house. With facial recognition features, Kuri is actually capable of recognising people and even judging their moods. It alerts you when it spots unknown people or something amiss in the house. A 4-microphone array helps Kuri react to voice commands and noises. Its speakers can also entertain you with music, engage your kids with stories, or help you catch up on news or podcasts. Kuri is also able to learn general patterns of behaviour. So it can wake you up if you are late for work! With easily programmable tasks Kuri is a cute robot that helps, inspires and and ‘IF This Then That’ (IFTTT) aukeeps an eye on your house (Courtesy: tomation capabilities, Kuri connects Mayfield Robotics) with your smarthome systems. It has a long battery life, and even when it is out of sap, it just goes for a nap near its charging pad and recharges automatically! Above all else, Kuri can connect with the family emotionally. It reacts to pats on the head, speaks to you in a very cute robot language, keeps your kids company and even plays with your pets. So it is a security, helper and friend, all in one! Company: Mayfield Robotics; Country: United States of America; Website: https://www.heykuri.com/

LyfieEye: Capture 360-degree content with your mobile phone There is nothing new about 360-degree cameras, but one that is tiny and works with your Android smartphone is truly awesome! eCapture Technologies claims that its LyfieEye is the world’s smallest 360-degree video camera for Android phones. This affordable device plugs directly into your phone and enables you to record, play and share life-like spherical 360-degree videos and photos. The content can be shared instantly on Facebook 360 or YouTube 360. The camera can also be synced to a virtual reality headset, so you can relive your special moments. Inside the 18gm camera are two Super-FishEye lenses, each providing a 180-degree field-of-view. Image sensors help to capture whatever the lenses are pointing at. The camera records videos at 30 frames per second. Thanks to its patented chip, Lywww.eFYmag.com

LyfieEye is a tiny 360-degree camera that plugs into your Android phone (Courtesy: eCapture Technologies)

fieEye can instantly flatten and stitch together images and videos captured by the two lenses and sensors. The recording is then transmitted in real time to your smartphone via the plug-in connector. Plug-in connection ensures that your 360-degree camera works reliably without depending on Wi-Fi or Bluetooth connections. LyfieEye is available at an MRP of $149. Company: eCapture Technologies Inc.; Country: United States of America; Website: https://www.lyfie.com/

U by Moen: Wi-Fi connected shower controller Your morning shower can make or break your day! Sometimes the water is extremely hot or cold if somebody else in the family has changed the mixer settings, or the hot water takes too long to reach the faucet. This makes you upset, especially if you are in a hurry to get ready for the office. For such situations, U by Moen smart shower is definitely a product worth investing in. The kit includes a digital thermostatic valve, a Wi-Fi enabled digital controller and a mobile app, which lets you remotely set up your bath from anywhere in your home. The kit can be

Customise your bath with U by Moen smart shower

used to control up to four water outlets, depending on your existing shower setup. The valve setup is supposedly do-it-yourself, but early adopters say that it is a bit complicated and requires a plumber’s services. Once you set up the system, you can customise elements of your bath such as temperature, time and outlet selection using the controller or mobile app. You can configure up to twelve presets, so all your family can get their bath the way they want it. When it is time for your bath, the system drains the cold water remaining in the pipes, heats the water to the required temperature, pauses the water flow, and waits for you to step in and enjoy your bath. For people with real need or love for luxury, the smart shower is a worthy investment. It can also be useful in boutique hotels, which can save preferences of their regular clients to boost customer satisfaction. Company: Moen; Country: United States of America; Website: http:// www.moen.com/u electronics For You | June 2017

87

IntervIew Robotics

“Robots Made In India Cost LoweR Than Their Imported Counterparts” RameshwaR singh ThakuR

chaiRman, TaL manufacTuRing soLuTions

TAL Manufacturing Solutions, a TATA Enterprise and a wholly owned subsidiary of TATA Motors, is currently the only manufacturer of commercial robots in India. It is also a notable producer of aerospace components. In a conversation with Paromik Chakraborty from Electronics For You, TAL chairman Rameshwar Singh Thakur talks about how robotics can benefit the manufacturing sector, the misconception about automation, and TAL’s robot BRABO

Q. what are the benefits of using robotics in manufacturing? A. The biggest intention of automation is to use robots in the dull, dirty and dangerous areas of work alongside multiplying the productivity and quality of yield. It is natural for human concentration and effort to fluctuate during strenuous, repetitive and risky work. Moreover, for human labour, you have sickness, marriage and other kinds of leaves of absence that might affect your delivery deadline, quantity of yield and quality of product. It is important to understand that robots are a better proposition to take up tasks that require continuous precision and consistency, especially in hazardous environments. The aim is to assign robots to routine jobs while humans take up the high-level work. Till now, we have 60 robots in the industry and all of our customers have reported at least a 25 per cent jump in productivity coupled with improved quality.

Q. could you please explain with some example? A. There is an auto components manufacturer in Satara, Pune, which supplies parts to motorcycle manufacturers. Its plant had three foreignmade robots. We replaced all three 88

June 2017 | electronics For You

A. Robots made in India cost lower than their imported counterparts. For instance, our robots are 40 per cent cheaper than those procured from overseas, as we manufacture every component, except the motor, domestically. Thus, our products can be serviced quickly, too. Today, our robots cost between ` 400,000 and ` 700,000 and target the SME sector.

the most competitive country in the world. Hyundai came into the industry much later than other automobile companies and today, it is rubbing shoulders with all of them. The Europeans too have more than 300 robots per 10,000 workers. People may say that Korea and Europe have less population and shortage of labour. But China does not have shortage of labour, and still it is going tremendously strong with the use of robots. There is a mindset that robots will take away jobs. The truth is that if you are not competitive, you won’t be able to get a job anyhow. Robots are making people competitive.

Q. what’s the impact of robotics adoption on employment worldwide?

Q. coming to your recently launched robot BRaBo, what all it can do?

A. Let’s take the example of China. China has a large population but they do not have the problem of unemployment. They use 49 robots for every 10,000 workers. Compared to that, India has a record of only two robots for every 10,000 workers. Despite having more robots and a large population, China has no problem with jobs. We can also look at countries like Korea, which uses 600 robots for every 10,000 workers. Korea is

A. We have provided functions of dispensing, pick-and-place, machine tending, press tending, optical inspection and so on in BRABO. We are also able to integrate the product with anything. The things BRABO can’t do right now are painting and welding. While we have no plans of any upgrade in BRABO as far as painting is concerned, it will be continually upgraded in other areas. The upgrades will all be customer driven.

of them with two of our robots. The manufacturer achieved a 23 per cent increase in productivity after the replacement.

Q. isn’t robotics an expensive proposition?

www.eFYmag.com

IntervIew Industrial Automation

“Collaborative robots Help Create Jobs, Not Steal Them” PradeeP david general manager, india & Sri lanka, UniverSal robotS

At EFY Conferences in March this year, there was a talk about integration of a new generation of robot workers in the labour workforce in order to increase productivity and efficiency. Known as ‘Cobots,’ short for Collaborative Robots, these robots are a combination of industrial robotics and automation. Pradeep David, general manager, India & Sri Lanka, Universal Robots, speaks to Dilin Anand from EFY about the role of cobots in production floors and the impact on human workforce

Q. Unmanned manufacturing—does it really make sense now that robots have arrived? A. The idea is for humans and robots to be inter-dependent and achieve what each of them does best, safely. There are a few things requiring human ingenuity that are best done manually, whereas accuracy, precision and repetitive mass production at higher efficiency are best taken care of by robots. Cobots provide an added incentive to the workforce as consistent quality production is possible with human supervision.

Q. What sets cobots apart? A. Collaborative robots are lightweight, cost-effective, safe and easy to operate. The human-robot collaboration is different from that between traditional robots. With cobots, the days of hiring an expensive external consulting team every time a robot has to be programmed are over. The new reality is that even operators with no programming experience can quickly program the cobot arms (using intuitive, 3D visualisation).

Q. How easy are they to operate? A. All the has to do is move the robot arm to the desired waypoints or touch arrow keys on an 90

June 2017 | electronics For You

easy-to-use touchscreen tablet. The cobot arms come with an average payback time of 195 days. That’s the fastest in the industry—quite simply because they are void of all the added costs traditionally associated with automation, such as external programming resources and shielded work cells. So automation can no longer be considered out of reach now.

Q. Which applications stand to benefit most? A. Production setups today often need to be flexible and agile in order to meet changing market demands and stay competitive. The lightweight cobot arms can be easily moved and re-deployed to new processes, enabling s to automate virtually any manual task, including those with small batches or fast changeovers. They can work right next to humans without any safety guard, thus making robotics technology accessible to all levels of the industry. Cobots are ideally suited for jobs that human operators find repetitive and dull.

Q. What differentiates them from conventional industrial and robotic arms? A. Industrial robots are caged to keep humans protected. Service

robots are meant to safely leave the cage while doing tasks for humans. Collaborative robots, on the other hand, are ‘force limited’ robots—their built-in technology allows them to work safely alongside humans (subject to safety assessments of the application). So these work more as a ‘worker’s assistant’ in the form of a ‘portable tool,’ unlike traditional industrial robots. Collaborative robots come in all sizes and shapes, and have integrated sensors and soft and rounded surface to ensure safety of human workforce and reduce the risk of damage due to impact, pinching and crushing.

Q. there are fears that robots will take away human jobs. With the rise of cobots, what is the future like? A. Instead of replacing human workforce and causing job losses, collaborative robots help companies expand operations, thereby creating jobs and many more opportunities at supervision level. A human-machine study conducted by MIT researchers at a BMW factory has found that teams comprising humans and robots collaborating efficiently can be around 85 per cent more productive than teams having either humans or robots alone. www.eFYmag.com

smart world

1. BharatQR Code: Not Just For Payments 92

3. 360-Degree Videos To Lower Costs

2. How About WhatsApp For Your Business 93

4. How Big Data Analytics Yields Big Gains 97

95

BharatQR Code: Not Just For Payments BharatQR Code was launched earlier this year (Image courtesy: thetechportal.com)

 BharatQR code is a transformative step towards easing and encouraging cashless payments  BharatQR is multi-purpose and can be applied in various facets of a business

E

nterprises in India are finally waking up to the QR code technology. Taking a leaf from other countries, the Indian government has launched a common QR code, BharatQR, with the aim to ‘reform’ the payment systems in the country. “BharatQR code is a transformative step towards easing and encouraging cashless payments. It makes digital payments far cheaper, and its key benefits like interoperability and seamless transactions make it extremely attractive for both merchants and customers,” shares Kishore Biyani, CEO of Future Group. “At Future Group, we are experimenting with the technology with the objective to roll it out in our stores in the near future. However, we believe there will be even more innovation in the payments area in 92

June 2017 | electronics For You

the near future that will make digital transactions even more intuitive, easy and accessible for every segment of Indian citizens. BharatQR code is a big step towards this direction,” Biyani adds. While the government might have wanted to only change the payment mechanisms in the country, this revolutionary technology is actually multi-purpose and can be applied in various facets of a business.

Increase sales No retailer can ignore the fact that most customers today first go online to do their research (for reviews, best price and quality of products) before buying a product. And not all are willing to come to the physical store to shop what they want. With the infrastructure for a common QR Code already

laid down, companies can generate their code on BharatQR and display it on their online store. When customers visit their online sites to see what is available in their stores, they can order and use the QR code to scan the payment. This allows companies to sell the stocks available in their brick-and-mortar stores, thus substantially increasing their sales. Back in 2011, Verizon achieved a 200 per cent increase in its sales using QR codes. The company simply displayed QR codes instores that enabled customers to enter a competition. Customers were required to share the brand’s ments on social media, and if their friend bought a product, they would get a smartphone. According to reports, the company clocked in an additional $35,000 www.eFYmag.com

smart world in sales while the investment was just $1000. Another way to increase sales is to display the BharatQR code in print ments or billboards, which takes customers directly to the company’s website once they scan the code. The company can then process customers’ orders and take the payment using the same QR code, thus saving both time and money in the process. A Vancouver-based coffee roaster, Ethical Bean, started using QR codes to attract customers. The company displayed QR codes in its train ads, which, when scanned, took customers directly to the cafe’s menu. This allowed customers to place their orders on the go, thus eliminating the waiting time while helping the cafe clock-in higher sales. Lloyd Bernhardt of Ethical Bean told Wall St Journal that the technology helped them double their business.

Cut unnecessary costs For any retailer, installing a pointof-sale (PoS) machine is both a capital expenditure as well as operational expenditure. And even though it makes up a little less than 10 per cent of the total operating cost for a retailer, every expenditure eats into the profit-pie of the company. A PoS machine has intangible costs associated with it. It can cost up to ` 15,000 and usually comes with a monthly maintenance charge, performance charge along with merchant discount charges (MDR). The MDR for a retailer varies from 0.25 per cent to 0.50 per cent for transactions up to ` 1000 and ` 2000, respectively. For transactions made from credit cards, there is no cap and sometimes it can go up to 2 per cent of the transaction value. For a transaction value above ` 2000, the MDR is one per cent of the total bill amount. “While PoS machines may not www.eFYmag.com

tional campaign for ESPN College Football around the Bowl Championship Series in 2013. The brand managed to engage a substantial number of customers by using QR codes smartly.

Seamless customer experience

Kishore Biyani, CEO of Future Group (Image courtesy: India Retail Forum)

contribute significantly to the operating costs of a large retail chain, a common QR code will definitely help medium- and small-scale retailers become more efficient. They just need to get a printout and stick it at the counter,” says fintech expert Abhishant Pant. Besides, the on-site maintenance of PoS machines is high and is a running cost throughout the lifespan of the machine. For instance, the rental for a PoS machine, depending on its type, is anywhere between ` 400 and ` 1500. Also, some PoS machines have performance charges ranging from ` 250 to ` 500. However, using BharatQR, retailers can eliminate this expense. The cost involved is negligible as it only requires the retailer to display the QR code on a piece of paper. The MDR is also lower. Apart from this, retailers need not pay any third-party fee to the payment gateway as BharatQR code is a common code that can accept payments from all three payment gateways—Visa, MasterCard and RuPay.

Effective marketing American fast-food restaurant Taco Bell managed to garner over 200,000 scans using QR codes in its promo-

The waiting time for a customer is a key factor in determining the footfall of a retail store. The longer the queue, the more likely the customer is to avoid the store. So, retailers are devising ways to reduce the waiting time at the queue in order to give customers a more seamless experience. BharatQR seems to address this concern to some extent. With PoS machines, consumers have to swipe their cards, enter their PIN and wait for the machine to generate receipts. While in the case of BharatQR, all that customers need to do is to scan the QR code before them and enter the PIN, to get the payment done. —Purba Das, senior business journalist, EFY

How About WhatsApp For Your Business  Instant messenger WhatsApp is a great way to do business. Nearly no-cost, no-frill tool, if used smartly, it can change the way communication is done in India and across the globe

B

ack in the 1950s, communication as a concept was not much popular across the globe. The regular TV, radio and print were probably the only electronics For You | June 2017

93

smart world

media. Cut to 2017, the world is literally running on communication. Be it a job application, a political campaign, a promotional campaign or anything under the sun, communication is pivotal to their success. And with a gamut of brands trying to communicate with their consumers on a daily basis, personalisation is imperative. So when WhatsApp came into existence in 2009, they took no time in exploring and exploiting its potential to make their businesses more efficient and yielding.

Sales and marketing There is no business in this world that does not involve sales and marketing. And definitely, no company that is not looking for effective ways to either up their sales or create a lasting brand recall in the minds of their consumers. WhatsApp offers this opportunity. Brands are rapidly using this application to connect with their consumers better. For instance, when Delhi’s Cafe TC (formerly known as Turquoise Cottage) was revamped, the owners wanted to promote its new look, new menu and happy hours amongst young target audience. However, they did not want run-of-the-mill advertising for their restaurant, and in their shoestring 94

June 2017 | electronics For You

budget they could not have afforded fancy campaigns. Hence, their digital agency Dentsu WebChutney came up with a quirky idea to use WhatsApp for its digital campaign. The agency modified WhatsApp’s API to change their status and profile pictures to a running ticker and GIFs. The agency then used all the content—the restaurant’s new logo, new menu and new offers among others—on the ticker it had created. The brand and its agency then promoted the campaign on social media sites such as Twitter and Facebook and asked the target audience to add their WhatsApp number to check out the restaurant’s new menu and exciting offers. The result was very impressive. The enquiries for the restaurant went up by a whopping 117 per cent while the sales surged by 28 per cent for the month. Globally, only a few brands have tried using WhatsApp for marketing. Back in 2012, Toyota Spain partnered with the interactive agency Medialabs to develop an innovative online marketing campaign using WhatsApp. The brand invited s to put ‘Hibrizado’ (Hybridized) as their WhatsApp status, after ing themselves on the brand’s website for a week. Once done, s could enter a competition to win iPhones in a lucky draw. Toyota Spain had aimed at highlighting the benefit of their range of models with hybrid technology. Through the campaign, the brand clocked nearly 35,000 visits and over 6500 ed s in just one month.

Customer services Customer services play a crucial role in the success story of an enterprise. A shoddy customer desk can tarnish the brand’s image, which, in turn, can have an adverse impact on the sales in the long run. Besides, customers these days prefer a one-on-one chat with a brand. WhatsApp offers exactly that—a direct personal chat.

“Consumers expect to engage, enquire and request service whenever, wherever, and via whatever channel suits their fancy. But addressing this paradigm shift in a positive way that delivers consistent company-customer interactions across channels and throughout every engagement is something that has an impact across the entire enterprise,” says Jim Freeze, senior vice-president, Aspect Software, in his blog “WhatsApp—Is It “What’s Next” in Customer Service?” His company, Aspect Software, offers customer interaction management, workforce optimisation and backoffice solutions to enterprises. Pune-based real estate developer VastuShodh partnered with advertising agency Setu to launch an integrated promotional campaign ‘Homes for all’ for its new projects. The company opted for both traditional and new media such as TV, print, out-of-home and social media among others to its projects. However, it used WhatsApp to connect with customers for business enquiries. The real estate developer provided a WhatsApp number in its ments, so that potential home buyers could enquire about the projects. It claims to have received over 800 business enquiries from WhatsApp alone while creating a strong database of genuine customers. Similarly, Rishi Gangoly of a firm called Webdudes conducts WhatsAppinar (WhatsApp versions of Webinars) for its products and services. Webdudes provides end-to-end solutions for designs and communications to enterprises. “We can have 50 people participate in the WhatsAppinar and get information about our products/ services for virtually no cost. These are interactive, so participants can engage. People who get busy during a WhatsAppinar can take calls without disturbing the peace and then www.eFYmag.com

smart world get back on to the WhatsAppinar and catch up on existing questions and answers that other participants on the group may have enquired about,” informs Gangoly. The company invites prospective clients to the WhatsAppinar by email. Once ed, s get a link to the WhatsAppinar 24 hours prior to the WhatsAppinar, following which the sessions begin. Once the sessions are over, s can post their queries or suggestions. Freeze of Aspect Software, however, notes that for brands to use WhatsApp effectively, they need to integrate it efficiently with their other channels. “Adding a service channel that consumers use and trust makes sense to increase engagement and provide service on the of your customers. But if that channel is not fully integrated with access to customer data, reporting resources and the expertise of the centre, it will hardly deliver an optimal experience. By applying centre best practices to mobile customer engagements, organisations can realise better value from the channel at a lower cost, without sacrificing the disciplines and service levels that are making the centre the de facto centre of the customer experience.”

Internal communication Considering the privacy that WhatsApp offers, most enterprises often use it for internal communications. Be it discussing a project to implementing employee engagement initiative, WhatsApp serves as a perfect platform to communicate. For instance, companies such as insurance web aggregator Policy Bazaar use WhatsApp for internal communications. The company has separate WhatsApp groups of each business unit as well as the management, wherein important information and updates are shared with everyone. www.eFYmag.com

“For us, it looked like the perfect tool to keep everyone in the loop and help managers reach their team at any moment of their workday. It is also a good way to maintain these employees’ engagement to their company by offering them the same information as the ones who can access the corporate website or intranet,” says Alexandra Guimaraes, an HR expert in her blog “Should you use WhatsApp for your internal communications?.” It is interesting to note that while WhatsApp is great tool for companies to do their business efficiently, unfortunately it has not yet been explored much by India Inc. However, once they realise the value of this simple instant messenger, WhatsApp could be the next big thing in the communication space in the country. —Purba Das

360-Degree Videos To Lower Costs  Use of 360-degree videos for corporate training is best suited to large-scale enterprises that often organise training sessions for their employees across the country or overseas

E

nterprises are often scouting for the next best technology for their businesses. After all, managers want their companies to become profitable—be it by costcutting or implementing systems

that help them connect with their customers better. That’s exactly what the creative agency behind Flipkart’s Treasure Hunt campaign had on its mind. To promote Flipkart’s annual Big Billion Day sale, Dentsu Webchutney, the creative agency that executed the campaign, used a 360-degree technology to give customers a tour of Flipkart’s head office in Bengaluru where customers had to hunt and earn their treasures, to get discounts on the products listed on the website. “For us, engaging with our s with a compelling story is as important as sales. We constantly like to push the frontiers—be it with our service, technology or consumer engagements,” Shoumyan Biswas, vice president-marketing at Flipkart, had said some time back. This is just one example where 360-degree cameras have been used to get some exceptional response. We explore here other benefits from using this technology.

Lower your surveillance cost The most important application of 360-degree cameras is security and surveillance. Nowadays, most companies install surveillance cameras to ensure safety and security of their assets. And the need for upping the security will only grow more. Small-scale enterprises need to spend at least ` 100,000, while medium-scale or large enterprises spend ` 1 million to ` 1.5 million on the entire surveillance solution for their office premises. Usually, cameras installed are the traditional pan-tilt-zoom (PTZ) type that have a narrow field of view and require additional components to monitor an entire area, which makes them more susceptible to mechanical failures. Also, these cameras have a higher chance of missing out a critical activity as they cannot capture an all-round electronics For You | June 2017

95

smart world view of a particular area at the same time. With 360-degree surveillance cameras, that is not the case. 360-degree cameras use a fish-eye lens to capture an all-round view of the field under surveillance. Thus these cameras can substantially bring down the expense incurred on surveillance, as enterprises would require fewer number of cameras to be installed over an area. For instance, if an area within the office premises required two PTZ cameras to get a full view of

interactive as they break the monotony of flat, plain-vanilla videos. Besides, 360-degree videos offer a real-time experience to trainees, giving them a better understanding of the subject under discussion. These videos are also considered to be cost-efficient as they eliminate the need for an in-person trainer. Use of 360-degree videos for corporate training is best suited to large-scale enterprises that often organise training sessions for their employees across the country or overseas.

set up a full-fledged infrastructure that includes building and staff among others. Their trained doctors can do the job through 360degree videos.

Enhance brand value Let’s go back to Facebook’s Treasure Hunt campaign that used 360-degree videos to engage with customers for their annual Big Billion Day sales. The campaign managed to increase the retailer’s Instagram base by a whopping 350 per cent while it garnered nearly 750,000 views and over 21,000 comments within the first two days itself. Other brands too have tried the immersive video technology and achieved success in their initiatives.

Better customer connect

A 360-degree camera being used for surveillance (Image courtesy: YouTube)

the area, it would now require only one panoramic camera for monitoring the entire area. These cameras are especially useful for logistics, retail, entertainment and other such enterprises that see movements of a large number of goods and people.

Corporate training While use of 360-degree videos or immersive videos is not new, corporates are exploiting their potential in other applications too. One such application is corporate training. According to experts, corporate trainings imparted using immersive videos turn out to be far more 96

June 2017 | electronics For You

Reduce cost of investment For healthcare sector, 360-degree videos are a definite boon. In a country such as India where people cannot avail even the basic medical care, 360-degree videos can help healthcare providers save many lives as services become costefficient. For instance, in a tier 2 or tier 3 city, a hospital chain need not set up an entire hospital but rather partner with a local hospital, provide it with the requisite equipment and train as well as supervise local doctors to diagnose and even carry out surgeries. Thus, for hospital chains, 360-degree videos completely eliminate the need to

Companies can use 360-degree videos for their products on display online, giving customers a virtual touch-and-feel experience. This helps not only give customers a better understanding of the product but also win their trust. This is especially true in the case of highvalue items such as diamonds and smartphones. Similarly, companies that need to invest in large distributor networks for their products can use this technology to help customers select the best option. For instance, auto companies are increasingly suggesting their dealer networks to install 360-degree cameras in their stores in order to ensure that best facilities and services are offered to their customers.

Is 360 degree the way ahead? Definitely. Any technology that helps enterprises enhance their business is here to stay. So if you belong to the old school of triedand-tested ideology, you may just be missing out on a great opportunity to scale up your revenues. —Purba Das www.eFYmag.com

smart world

How Big Data Analytics Yields Big Gains  Data collected from surveys, forms, reports and social media can be analysed using big data analytics to understand the specific needs of the consumers

R

ed Roof Inn, a budget hotel chain in the U.S., could have never imagined that it would be able to clock in a 10 per cent increase in its business when it used big data analytics to attract its potential customers in the winters of 2014. The hotel chain realised that most of its customers in bad-weather-affected locations would be using mobile devices to search for hotels. Hence the brand created targeted campaigns for such customers, which resulted in a surge in its footfall. “The key for any company that wants to successfully use big data is gaining the right information that delivers knowledge and gives businesses the power to gain a competitive edge. And this can only be done by identifying and selecting from different types of big data analytics,” says Naveen Joshi, director at Allerin Tech, in his LinkedIn post. Big data analytics offers much more than just target marketing. If implemented wisely, it can help a business grow by leaps and bounds.

Revenues maximised using existing resources Imagine knowing where the demand for your product or service is coming www.eFYmag.com

from and then using existing resources with optimal efficiency to ensure that you cater to that demand right on time. In fact, a few companies have started using big data analytics to do exactly that. For instance, hospitality giant Starwood Hotels and Resorts managed to increase its revenues-per-room by nearly 5 per cent by using big data in 2015. They used this technology to predict the demand as well as recalibrate room fares in order to get a higher footfall. The company claims that its demand forecasting has improved by 20 per cent since 2015. Starwood used predictive analytics to understand its customers from their transactional data (number of bookings made, cancellations and others) and combined it with external semistructured data such as weather reports for each of its locations. Since then, the company has been using analytical software to optimise its efficiency. Similarly, another hotel major Marriott uses big data to enhance yield management of its rooms, thus maximising revenues. The company uses predictive analytics to predict the demand and accordingly make pricing decisions. Experts project that predictive analytics could really be a game changer if used wisely. Implementing big data analytics while using existing resources is a smart way to up your revenues without much hassle.

Reduced operational costs Every business strives to be efficient when it comes to operational costs. And companies in the West have been adopting big data analytics for this purpose. They implement big data analytics in their internal processes to gather data about their operational costs and use this data to downsize unnecessary expenses. For instance, since 2012, Intel has been using big data to launch its chips in the market faster. According to reports, the company uses predictive analytics to understand and

analyse historical data gathered from pre-release chips in order to cut down the time spent on quality checks of its chips. Instead of running every chip through 19,000 tests, the company selects certain chips for those tests to ensure the quality of its chips. If reports are to be believed, the company saved $3 million in 2012. Similarly, retail giant Tesco used big data analytics to find out the exact temperature required for storage in its refrigerators. The company found out that most of its stores in Ireland were running refrigerators at a much lower temperature than needed. Thus, by understanding the fault and taking corrective step, the company was able to successfully reduce its refrigeration cooling costs by 20 per cent. It is only a matter of time before big data analytics becomes imperative to run a business.

Enhanced customer engagement Companies have to constantly innovate in order to expand their consumer base. And this can only happen when they have a clear understanding of their consumers’ needs. Industries such as banking and tourism can also turn to big data analytics to deliver customised products to their customers.

Way forward The biggest challenge that continues to haunt companies when it comes to big data analytics is the quantum of investment required to implement the technology. And while it is true that this technology is pricier than traditional ways of data analysis, the benefits of implementing it far outweigh the cost. To summarise, big data analytics—be it descriptive, predictive or prescriptive—is inevitable for a business. The sooner India Inc understands it, the better! —Purba Das If you want to share with EFY readers case studies of technology being applied to reduce costs, increase revenue, go green, please us at [email protected] electronics For You | June 2017

97

Industry News Make in india

AI could replace doctors and lawyers On The Move in five years Schneider Electric appoints By 2022, smart machines and robots may replace highly trained professionals in tasks within medicine, law and IT—forecasts leading research and advisory company Gartner. “The economics of artificial intelligence (AI) and machine learning will lead to many tasks performed by professionals today becoming low-cost utilities,” says Stephen Prentice, vice president and Gartner Fellow. “AI’s effects on different industries will force enterprises to adjust their business strategy. Many competitive, high-margin industries will become more like utilities as AI turns complex work into

a metered service that the enterprise pays for, like electricity.” The effects of AI on an enterprise will depend on its industry, business, organisation and customers. Law and financial services are among the industries where jobs could be automated. However, while AI will hit employment in some industries, many others will benefit as automation of routine and repetitive tasks will leave more time for the existing workforce to improve service levels, handle more challenging aspects of the role and even ease stress levels in some high-pressure environments.

OTT to sur broadcast TV Within the next five years, livelinear streaming over-the-top (OTT) video viewership will sur traditional broadcast TV viewership, according to close to 70 per cent of respondents in the 2017 OTT Video Services Study by Level 3 Communications, Streaming Media and Unisphere Research. This annual survey on OTT trends was based on respons-

es from almost 500 media industry professionals. “The maturation of OTT video delivery reflects key changes in the industry, including fewer bandwidth constraints and the important role live-event and live-linear OTT distribution will play as we transition from VODonly services to those that more closely mimick traditional broad-

Swaminathan as VP

Schneider Electric has appointed Venkatraman Swaminathan as the vice-president, IT Division - India Zone. Swaminathan takes over from Nikhil Pathak, who has moved to a new role as the vice president-strategy deployment and operations, ROWIT Business at Schneider Electric. In his new role, Swaminathan will be responsible for driving the growth of the overall IT business in India. With over 26 years of experience in the field of sales, service and marketing, he brings a wealth of experience in the field, having evolved business-critical solutions for spaces like data centre switch rooms and test laboratories.

cast delivery,” said Tim Siglin, contributing editor for Streaming Media magazine and principal report author. The report finds that the OTT industry is booming. About half of respondents anticipate OTT revenues to grow 30 to 50 per cent from 2016 to 2017. Over the next three years, OTT-related services will for more than a quarter of their overall business, according to nearly two-thirds of respondents.

New technologies drive double-digit growth in connected homes The connected homes market will grow beyond $150 billion, with revenues growing at 14.9 per cent in 2017. Technologies such as Internet of Things (IoT), Big Data, data analytics and the cloud are propelling doubledigit growth and market expansion in connected homes and light-emitting diode (LED) lighting sectors—finds a 98

June 2017 | electronics For You

study by Frost & Sullivan. Smart thermostats, Internet protocol cameras, smart door locks, IoT-enabled LED lighting and personalised mobile health services will see strong growth. Data analytics will penetrate beyond energy management, to equipment management, people management

and supply chain. The cloud’s full potential in energy management will be realised by developing a centralised data repository for customers to access their building’s energy data, while connected HVAC systems in the commercial segment will augment predictive maintenance and remote diagnostics capabilities. www.eFYmag.com

Industry news Calendar of Forthcoming Electronics Fairs/Exhibitions/Seminars/Events Name, Date and Venue

Topics

address for details

E3 2017 June 13-15, 2017 Los Angeles, California, USA

Annual video game conference and show

Entertainment Software Association Website: www.e3expo.com

TechXLR8 June 13-15, 2017 ExCeL, London, UK

Festival of technology for a journey between networks, tech and consumer services

TechXLR8 Website: https://tmt.knect365.com

Internet of Things World Europe June 13-15, 2017 ExCel, London

Platform to discuss adoption of Internet of Things (IoT) across consumer verticals

KNect365 Website: https://tmt.knect365.com/ iot-world-europe/

Cyber Security Summit 2017 June 15, 2017 Hyatt Regency, Mumbai

It will bring together 120+ IT security leaders and policy makers from across industries

[email protected]

SES 2017 July 6-7, 2017 Bangalore International Exhibition Centre

8th edition of strategic electronics summit with focus on military, aerospace, homeland security and naval systems

ELCINA [email protected]

KE Show 2017 July 11-14, 2017 Huaqiao International Expo Centre, Kunshan, Jiangsu Province, China

Event on intelligent automation and robotics

Taiwan Electrical and Electronic Manufacturers’ Association (TEEMA) Website: www.TEEMA.org.tw

IFA 2017 September 1-6, 2017 Berlin,

Meeting place for key retailers, buyers, and experts from the consumer electronics industry and the media

The German Association for Entertainment and Communications Electronics and Messe Berlin Website: http://b2b.ifa-berlin.com

World Robot Olympiad India National Championship 2017 September 9-10, 2017

Robotics competition in India for students between the age group of 9 to 25 years

India STEM Foundation, and National Council of Science and Museum. Website: https://indiastemfoundation. org/wroindia/

AI World 2017 September 18-20, 2017 Boston, USA

Artificial intelligence (AI) world conference and expo

AI World 2017 Website: https://aiworld.com

TechXLR8 Asia October 2-4, 2017 Marina Bay Sands, Singapore

Festival of technology that brings together Asia’s telco and tech ecosystem to discover technologies that are enabling and shaping our connected future

TechXLR8 Asia Website: https://tmt.knect365.com/ techxlr8-asia/

Taitronics Taipei October 11-14, 2017 Taipei, Taiwan

International trade fair for electronics and electronic components

Taiwan External Trade Development Council Website: http://www.taitra.org.tw/

Open Source India October 13-14, 2017 NIMHANS Convention Center, Bengaluru, India

Asia’s annual convention to celebrate open source computing

Open Source India Website: http://opensourceindia.in/ osidays/

Assocham Green Energy Summit November 9-12, 2017 Bangalore International Exhibition Centre

World expo and conference on renewable energy

ASSOCHAM India

Big Data London November 15-16, 2017 Olympia, London, UK

The UK’S largest data and analytics event

Big Data London Website: http://bigdataldn.com/

CeBIT INDIA 2017 November 30-December 2, 2017 White Orchid Convention Centre, Hebbal, Bengaluru

Digital innovation show

Hannover Milano Fairs India Pvt Ltd Website: www.cebit-india.com

ELEXCON 2017 Annual expo for the electronics industry based December 21-23, 2017 in Shenzhen Shenzhen Convention & Exhibition Center, China

ELEXCON 2017 Website: www.elexcon.com/elexcon/en

CES 2018 January 9-12, 2018 Las Vegas, Nevada, USA

World’s gathering place for those who thrive on the business of consumer technologies and where next-generation innovations are introduced to the marketplace

Consumer Technology Association Website: www.ces.tech

NEPCON Japan January 17-19, 2018 Tokyo Big Sight, Japan

Asia’s leading exhibition for electronics design, R&D and manufacturing technology

Reed Exhibitions Japan Ltd Website: http://www.nepconjapan.jp/en/

Electronics West February 6-8, 2018 Anaheim Convention Center, California, USA

North America’s largest annual design and manufacturing event

Electronics West Website: http://electronicswest. designnews.com/

Embedded World 2018 February 27-March 1, 2018 Exhibition Centre, Nuremberg,

Trade fair for IT security for electronic systems

NürnbergMesse Website: www.embedded-world.de/en

Continued on next page... www.eFYmag.com

Smart home device sales to double in four years In four years, annual growth of smartphones sold will reduce from 30 per cent in 2014 to a mere four per cent, according to a new study by research and consulting firm Ovum. Meanwhile, the number of connected smart devices in use will nearly double from eight billion last year to 15 billion in 2021. This will lead smartphone majors like Apple and Samsung to seek new sources of growth in virtual reality, smart home and other remerging segments. Interactive audio speakers offer a more natural way than smartphones or tablets to control smart home devices. For example, Amazon’s Alexa already has more than 10,000 ‘skills’ that allow it to interact with a very large number of apps. However, smartphones will still be a critical component of driving smart homes, and by 2021 these will cement their position as the most popular consumer device of all time, with more than five billion in use.

Tech majors racing to invest in AI startups According to a report by CB Insights, leading technology firms are investing in AI in order to capitalise on the shifting digital market. Prominent among them are Alphabet, Intel, Microsoft, IBM and Amazon. In the last five years, Alphabet, the parent company of Google, has acquired as many as eleven companies closely associated with artificial intelligence. Google already uses RankBrain—an algorithm learning artificial intelligence system used in Google Search—for understanding queries and processing suitable responses. In 2016, Intel acquired Itseez, Nervana Systems and Movidius, on par with Apple’s acquisitions. It plans to develop semiconductors for AI apps. Microsoft has been investing in AI technology for about 25 years now. electronics For You | June 2017

99

Industry news

Snippets Telit unveils new IoT University

Telit, a global enabler of the Internet of Things (IoT), has launched Telit IoT University for IoT developers and integrators. Telit IoT University, part of the company’s IoT Know How suite of consultative services, teaches customers and partners how to succeed in launching IoT solutions that deliver measurable business outcomes. Located at Telit IoT Platforms division in Boca Raton, Florida, Telit IoT University’s curriculum includes courses and hands-on labs co-developed by Telit and Florida Atlantic University’s Institute for Sensing and Embedded Network Systems Engineering (I-SENSE).

Government initiatives, falling prices driving India’s LED lighting market Market intelligence firm 6Wresearch expects India’s LED lighting market to grow annually at 26.6 per cent during 2017-23, owing to government initiatives such as UJALA, Housing for All and Deendayal Upadhayaya Gram Jyoti Yojna. Missions such as Smart Cities would also drive the LED lighting market in the coming years. At the same time, LED prices have declined tremendously in last few years, leading to surge in penetration across the country.

Tata Power, Dell build India’s largest vertical solar farm Tata Power Solar, India’s largest integrated solar company, and Dell International Services India, have together built India’s largest vertical solar farm of 120kW capacity. The 45-metre-long structure built at Dell’s Bengaluru campus provides dual benefits of producing sustainable green energy and insulation of the building by blocking the south sun, thus helping to reduce power consumption. The solar farm, envisaged by Dell, produces enough energy to light-up its entire cafeteria and basement parking.

Hartek Group separates rooftop solar business The Hartek Group, a power sector player, has set up Hartek Solar as a separate rooftop business vertical with industry veteran Ravinder Shan at its helm as chief executive officer (CEO). Having commissioned 13.75MW rooftop solar plants across the country, the Hartek Group rooftop solar business division has a two per cent market share. Shan revealed that the immediate focus of Hartek Solar would be on commercial spaces and industries.

Anritsu LTE-Advanced RF conformance test system wins GCF certification Anritsu’s LTE-Advanced RF conformance test system ME7873LA has won the first GCF certification for the Cat-M1 RF conformance test as part of the Internet of Things (IoT) communications technology. Large domestic and international communications carriers are planning the 2017 introduction of Cat-M1 and NB-IoT technologies for use by mobile networks, ing the growth of the so-called ‘IoT Society.’ With GCF certification, the ME7873LA has more than 70 test cases for frequency bands in Japan, N. America and Europe. Continued from previous page...

Calendar of Forthcoming Electronics Fairs/Exhibitions/Seminars/Events Name, Date and Venue

Topics

address for details

ELCERAMA March 10-14, 2018 India Expo Mart, Greater Noida, Uttar Pradesh

Flagship showcase of the Indian electrical industry ecosystem and the largest congregation of power sector ecosystem in the geography; brings together the complete spectrum of solutions that powers the planet

IEEMA Website: http://elecrama.com, http://ieema.org

Hong Kong Electronics Fair (Autumn Edition) October 13-16, 2017 Hong Kong Convention and Exhibition Centre

World’s leading electronics marketplace

Hong Kong Trade Development Council Website: www.hkelectronicsfairae.com

Since this information is subject to change, all those interested are advised to ascertain the details from the organisers before making any commitment.

In December, it launched an AI fund to focus solely on “investing in AI companies focused on inclu100

June 2017 | electronics For You

sive growth and positive impact on society.” With the most AI patents among

its peers in 2015, IBM has invested billions of dollars in developing the Watson cognitive technology that analyses and interprets data including unstructured text, images and videos. In March this year, IBM and Salesforce announced that the two firms will be integrating their independent AI platforms. Amazon has already created buzz with Alexa voice-enabled assistant that recognises and responds to voice requests instantly. Last year, it had launched a video that presented a utopian high-tech grocery store without cashiers.

Solar park costs on the rise in India Solar parks in India are aimed at reducing the hurdles and providing a level-playing field to developers, whether domestic or foreign. While these parks have spurred record low tariffs in the recent times, market intelligence firm Mercom has found that solar park costs are actually increasing. Incomplete solar park infrastructure, exorbitant upfront fees and not-so-transparent yearly charges are all adding to project costs, with some developers calling for auctions outside of solar parks as they think it could be cheaper in some instances. Even though all solar parks come under the same policy, each solar park has a different fee structure, which is leading to variation in tariffs and project costs. In fact, two solar parks in the same state can have different fees. According to Mercom’s data, the solar park fee in India varies from ` 3.73 million per megawatt in Karnataka to ` 10.58 million per megawatt in Gujarat. Land application fees (non-refundable one-time), upfront development charges (one-time), annual leases of land, and annual operation and maintenance charges are the most common park fees or charges. However, developers are faced with differing stamp duties in each state and service taxes on all components. Hidden charges end up inflating project costs. www.eFYmag.com

make in india Market Survey

Strategic electronicS Set For Revamp

L Sudeshna Das is senior executive editor at EFY

arge-scale modernisation of the Indian defence forces is on the anvil, making strategic electronics (mainly comprising aerospace and defence electronics) a vibrant industry. India needs to modernise its aerospace and defence capital equipment base by addressing obsolescence as well as building additional capabilities through new capital acquisition. The new Defence Procurement Policy (DPP) and offset policies announced by the Indian government are a big step in that direction. The government has also made budgetary allocations for paramilitary and state police force requirements. All these developments have made India quite attractive for aerospace and defence electronics manufacturers.

Market opportunities As many as 80 per cent (Fig. 1) of the survey participants feel that the industry will see a slow but steady growth in the coming decade. There will be an exponential growth in both standalone systems (as part of platforms) as well as other systems at sub-system level. According to a report by consultancy firm Roland Berger, in partnership with the National Association of Software & Services Companies (NASSCOM) and India Electronics and Semiconductor Association (IESA), India’s aerospace and defence industry is expected to consume electronics worth $70-72 billion

Methodology This report is based on inputs from 24 senior professionals involved in the strategic electronics ecosystem. Responses were taken either directly or from their speeches at Defcon conference conducted by the Electronic Industries Association of India (ELCINA) in April this year. This sample group is a microcosmic representation of India’s strategic electronics ecosystem. Survey participants shared their insights on: 1. Growth of the Indian defence electronics industry 2. Demand-generating applications 3. Possibility of public-private partnership (PPP) or industry-academia partnership 4. Impact of policy amendments 5. Challenges A trend analysis was done on the basis of inputs. Results of the analysis are presented here. www.efymag.com

in the next decade, as the country rapidly modernises its military by embracing new technologies. Electronics acquired as part of platforms (at Tier 1 and Tier 2 levels) will be worth $53-54 billion. Demand for electronics worth $17-18 billion is expected from projects that are traditionally called system-of-systems (non-platform based) applications. Another growth opportunity is the potential to integrate into global value chains of original equipment manufacturers (initially leveraging the offset route and subsequently based on product quality, service delivery advantages and cost arbitrage).

Market demand In spite of huge opportunities, there is a demand-supply gap for indigenous components, products and solutions, most of which are still imported. Indian companies are not able to cater to requirements that involve low volumes, high technology and high investments. Areas that require electronic products and solutions in large volumes include: 1. Electronics manufacturing services, ‘build to print’ (which, in the context of defence production, includes firmware updates, test-jig development, manufacturing software development, component procurement, systems integration, software upgrades/enhancement, verification and validation, system deployment, hardware qualification, environmental stress screening, maintenance , etc), line replaceable units for Indian programmes like Light Combat Aircraft (LCA) and Light Combat Helicopter (LCH), drones, Kamov helicopters, etc 2. Rugged displays, PDAs and laptops for battlefield management system project 3. Electronic components, products and solutions used in vehicles for projects like the Futuristic Infantry Combat Vehicle and selfpropelled guns Most of the recent upgrades in military systems have a huge scope for electronics, as in the case of Jaguar, Ilyushin aircrafts, MIG 29 and T-90 tanks. In addition, new ventures electronics for you | June 2017

101

make in india like the LCA Tejas, HAL Advanced Medium Combat Aircraft, EADS CASA C-295 twin-turboprop tactical military transport aircraft production tly by Airbus Defence and Space and Tata Advanced Systems, single-engine aircraft production by SAAB (Gripen) with a yet-to-be-chosen Indian OEM, FICV project, missile development programme, new aircraft carriers, submarines and Marine Counter Measure Vessels all point to an increased focus on electronics. Critical avionics, nightvision devices, missile electronics, electronic warfare (ground-based and airborne) systems, software-defined radios, robotics, radars (of all kinds), UAV electronic navigation systems, weapon sensors and air defence systems are the much publicised favourites. According to experts, major demand-generating applications will be weapon control systems, avionics and digitalisation of displays. In addition, there is a demand for single-board computers, advanced PCB design technology, cables and harnesses to meet aerospace standards; a holistic security network and indigenously developed security devices; and innovative maintenance, repair and operational solutions to sustain and upgrade existing systems.

80%

20%

Slow but steady growth Fast growth

Fig. 1: Growth forecast for the Indian strategic electronics sector

70%

Offsets will play a key role

30%

Not sure

Fig. 2: Offsets-related opportunities

80%

20%

Need of the hour Not of immediate importance

Fig. 3: Exploring design in India to achieve ‘Make in India’ goals

Offsets to play a key role Seventy per cent of survey participants (Fig. 2) feel that offsets will be on the rise. However, emphasis will be mainly on embedded intelligence. So offsets will have to be defined in of service offerings rather than hardware, quite quickly. The remaining 30 per cent of respondents raised concerns about the increase in threshold limit for offsets, under DPP 2016, to ` 20 billion (around $305 million) from the earlier limit of ` 3 billion. This is in contrast to global trends in defence offset thresholds, which hover around $5-$15 million in many countries (including Israel, Malaysia, Turkey and UAE). There are fears that this increased threshold limit may lead to a situation 102

June 2017 | electronics for you

80%

Yes

20%

Maybe

Fig. 4: Is public-private partnership a success?

R&D and manufacturing. As many as eighty per cent (Fig. 3) respondents opined that this is the right time to explore ‘Make in India’ opportunities by deg in India for the country’s defence electronics needs. Last year, the defence sector got a policy boost through amendment of DPP 2016 to include a special category called ‘Buy (Indian-Indigenously Designed, Developed and Manufactured)’ or ‘Buy (Indian-IDDM).’ This category will get the highest priority in the defence procurement process, followed by ‘Buy and Make (Indian),’ ‘Buy and Make’ and ‘Buy (Global)’ in that order. Inclusion of this new procurement category will provide a greater thrust to the ‘Make in India’ initiative in defence electronics production, focusing on products and solutions designed in India. This would promote in-house design capabilities and higher localisation, strengthening the domestic industry, especially private sector, in defence electronics production. However, the responsibility to prove a design as indigenous rests with the manufacturer, while the final approval would come from the government. This, in turn, offers unique opportunities to companies that are operating or would like to enter the defence industry specifically in the areas of electronic system design and original design manufacturing. The government has already launched some programmes and schemes to facilitate R&D activities in this field. The Technology Development Fund (TDF) and the Army Design Bureau (ADB) are worth mentioning here.

where fewer import contracts will be eligible for offsets. This, in turn, will impact the domestic industry, particuPartnering for success larly parts and components manufacturers who got the existing offset policy According to survey participants, boost to enhance their capabilities. collaboration is the key to achieving self-reliance in the strategic electronics Focus on ‘design in India’ sector. There is a need for high standards in this industry, which can be fulSurvey participants feel that the Indian strategic electronics sector is still filled if the Indian industries complement each other rather than compete. far from achieving self-reliance due to the absence of integration between The TDF and the ADB can play an www.efymag.com

make in india important role to ensure success. According to survey participants, it will be prudent on the part of the government to work on the PPP model, where the technology, manpower and infrastructure available with the public-sector units (PSUs) can be leveraged in partnership with private enterprises to optimise and streamline defence projects towards quick completion. The PPP model will introduce ordinance factories to a more competitive corporate culture—a mindset that is required to pitch for contracts with the private industry. Ordinance factories have vast infrastructure but they lack efficiency. Therefore, in the PPP model, the required levels of efficiency can be assured through proper assimilation of technology from the private sector and efficient utilisation of the huge infrastructure available with ordinance factories. This could lead to a win-win situation for both the private sector, which is looking to

www.efymag.com

foray into defence manufacturing, and the public sector, which has a strong grip on defence procurements. The Armed Forces have also been engaging with academia to solve various technical challenges. For instance, IIT Bombay was involved in a project to reduce the cost of the artillery shell guidance system. There are many more such examples. Around 65 per cent of respondents (Fig. 5) were optimistic about this kind of partnership. Moreover, the right kind of collaboration can also help to resolve supply chain related issues.

Policies: The good, and the bad As many as 75 per cent survey participants find DPP 2016 to be effective enough, plugging all the loopholes of the previous policy (Fig. 6). The remaining 25 per cent feel that it can be effective only when: 1. Tangible benefits are realised by Indian SMEs and MSMEs.

2. Global OEMs partner with Indian companies with actual manufacturing capabilities, rather than those with just financial might, to make the concept of Make in India doable. 3. The government gives more importance to the Ministry of Defence’s 23 key ‘Make in India’ projects designed by the Indian PSUs and DRDO rather than allowing foreign OEMs to manufacture their material in India. 4. The policy gets simplified and a penalty is imposed on the defence procurement agencies for delays and cancellations. Sixty-five per cent survey participants believe that removal of the 49 per cent FDI limit in the defence sector has opened up avenues for significant investments, and given the much required impetus for many SMEs to get access to finance and technology (Fig. 7). The liberalised FDI regime permits up to 100 per cent foreign equity in the defence sector. Of this, up to 49 per cent is

electronics for you | June 2017

103

make in india Major contributors to this report 35% 65%

Effective Not effective enough

Fig. 5: Industry-academia partnerships

Effective

75%

25%

Not effective enough

Fig. 6: DPP 2016

35% 65%

It will bring significant investments It may not be successful

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Atul D. Rane, scientist-G, Technology Development Fund, DRDO Lt Gen. A.R. Prasad, VSM, ADC, signal officer-in-chief and senior colonel commandant Lt Gen. K.K. Aggarwal, SM, VSM, Comdt, MCEME K. Rambabu, scientist-H, Space System & Tech, RCI Air Commodore Krishna Iyengar Ravi, AVSM, VSM (Retd), and VP (strategic business), Kaynes Technology Col (Retd) K.V. Kuber Lakshmi Tiwari, scientist-F, DLRL M.M. Sreeram, ED, Samuha Aerospace Maheshkumar P. Jagtap, scientist-G, ANURAG (Advanced NUmerical Research and Analysis Group), a lab under DRDO Nikhil Kumar Jain, Addl GM (AFS), BEL-Hyderabad N. Ramachandran, MD, Mel Systems Wg Cdr (Retd) P.R.L. Prakash, GM, Avantel P. Sudhakar, OSD to DAE, and ex-CMD, ECIL Ramesh Avala, manager (design), Strategic Electronic Research Design Centre (SLRDC), HAL, Hyderabad Ravindra Kumar, scientist-F, DLRL Rudra Kumar Jadeja, CEO, Bharat Forge Limited (Hyd) Dr S. Guruprasad, DG - PC & SI, DRDO HQs Sanjoy Sarma, head-products and solutions, Tata Power SED Sreeram Srinivasan, CEO, Syrma Technologies Subhash Goyal, MD, Digital Circuits T.V. Prakash Rao, ex-director-ECS, DRDO Utpal Sen, head-strategic electronics division, ECIL V.S.N. Murthy, scientist-G, project director, Brahmos Aerospace V.V.R. Sastry, ex-CMD, BEL, and ex-ED, C-DOT

will not be promoted through t ventures with higher FDI. Fig. 7: Revamped FDI policy

allowed through the automatic route and beyond that through the Foreign Investment Promotion Board (FIPB). The remaining 35 per cent feel: 1. The FDI policy cannot become the only instrument to attract investment. 2. The FDI policy will be successful only if foreign OEMs can derive the necessary confidence in the local ecosystem from an operational point of view. 3. Foreign OEMs are less likely to create intellectual property in India. Therefore ‘design in India’ concept

What to expect Survey participants feel that following factors could derail growth of the strategic electronics industry: 1. Complex buying process 2. Delay in completing the tender process 3. Inadequate government funding for R&D projects 4. Inadequate provision for technical superiority to score over the lowest bidder 5. Absence of a level-playing field for the Indian private-sector players, especially SMEs, with respect to PSUs and overseas vendors

Key provisions of DPP 2016 1. 2. 3. 4. 5.

Inclusion of the ‘Buy (Indian-IDDM)’ procurement category Rules for higher (though flexible) indigenous content requirements Revamped procedures for ‘Make’ projects Institutionalisation of the RFP (Request for Proposal) process Introduction of the L1-T1 (selection process based on both the lowest bid (L1) and the best technical solution (T1)) methodology for selecting the final bidder 6. Increase in threshold limit for offsets

104

June 2017 | electronics for you

6. No self-reliance in critical technologies 7. Absence of indigenous product development capabilities The above issues can be successfully resolved by the Transfer of Technology (ToT) route for niche equipment, proper understanding of offset obligations and a methodology to handle the same, and understanding and execution of international collaboration and t venture models with emphasis on IPRs and patents. In addition, proper screening and approval systems need to be put in place as components used in strategic electronics systems require a high degree of reliability in extreme working environments. Realising this need, ELCINA is setting up a high-tech test centre for the defence electronics sector with the objective of providing easy and cost-effective test facilities to MSMEs and SMEs. The facility will be open to all defence electronics manufacturers and the excess capacity will be offered to manufacturers in other sectors of the electronics industry. www.efymag.com

New Products Make in india

test & measurement Thermal imaging camera FLIR Systems, Inc. has introduced the ETS320 thermal imaging solution for electronics testing in engineering benchtop environments. This product combines a highsensitivity thermal camera with an adjustable, hands-free table stand to provide consistent, non- thermal testing. By offering more than 76,000 points of temperature measurement, it offers the ability to monitor power consumption, detect hot spots, and identify potential points of failure during product development. This rechargeable, battery-powered unit can be connected to a PC running FLIR Tools software for detailed data analysis, recording and reporting. FLIR Systems, Inc. www.flir.in/ETS320

Gas sensor Toshniwal Sensing Devices has introduced CO2 gas sensor module CDM7160 from FIGARO, Japan. This sensor can be used in indoor air-quality control systems, including variableair volume systems and demandcontrolled ventilation systems. It uses a compact NDIR CO2 sensor, providing high accuracy at low power. Two detector elements inside the module make absolute measurement possible. The module comes with both a UART and I2C digital interface. Evaluation module EM7160 is also available for easy testing of the CDM7160. It provides CO2 measurement data output in serial data format via a USB connector, pulse-width modulated (PWM) output and continuous voltwww.efymag.com

age output (converted from the PWM output signal).

These feature VCO phase noise of -113dBc/Hz at 100kHz offset at 5GHz, with integrated rms jitter of just 97 fsec (1kHz to 20MHz).

Toshniwal Sensing Devices Pvt Ltd www.tsdpl.com

Analog Devices, Inc. www.analog.com

WLAN test software

Silicon-carbide diodes

National Instruments (NI) has introduced WLAN Test Toolkit 17.0 with for Draft 1.1 of the IEEE 802.11ax standard. Combined with NI’s second-generation vector signal transceiver, the toolkit s 802.11ax waveform generation and analysis for characterisation, validation and production test of products such as RF front-end components, wireless modules and devices. The toolkit helps s solve new access point test cases by generating signals that simulate multi environments, including per- impairments.

STMicroelectronics has introduced a full range of 2A-40A 1200V silicon-carbide (SiC) JBS (Junction Barrier Schottky) diodes. Using SiC technology, the new diodes are said to offer high switching efficiency, fast recovery and consistent temperature characteristics. Featuring a lower forward voltage, these are suited for cost-conscious applications including solar inverters, industrial motor drives, home appliances and power adaptors. At the same time, performance-oriented applications can use these for superior efficiency, low weight, small size or best thermal properties. The higher efficiency provided by lower forward voltage drop delivers important benefits for automotive equipment such as on-board battery chargers and charging stations for plug-in hybrid or electric vehicles as well.

National Instruments www.ni.com/80211ax/

Components RF synthesisers Analog Devices’ 13.6GHz ADF5356 wide-band synthesiser with integrated voltage-controlled oscillator (VCO) is suited for applications like wireless infrastructure, microwave point-to-point links, electronic test and measurement, and satellite terminals. The ADF4356 is a complimentary synthesiser product that operates to 6.8GHz and is comparable in performance. The new synthesisers generate RF outputs from 53.125MHz to 13.6GHz without gaps in frequency coverage.

STMicroelectronics www.st.com

Radar sensor chips Texas Instruments (TI) claims to have unveiled the smallest millimetre-wave radar sensor chips that can measure distances smaller than the width of human hair. Utilising CMOS process, the new chips integrate both analogue and digital functions. These find applications in autonomous driving, industrial robotics, medical devices, and factory and building automation. The sensors include five solutions across two families of 76 to 81GHz electronics for you | June 2017

105

New Products sensors with a complete end-to-end development platform. Texas Instruments www.ti.com

Intelligent power module Infineon has developed a highlyintegrated intelligent power module to control induction motors and permanent magnet synchronous motors with single-phase power factor correction (PFC) in variable-speed drives. The new module, CIPOS Mini, combines the PFC stage with a 3-phase inverter in one package. This helps to reduce system size and the bill of material. The additional PCB space gives room for further components. The power module is available in inverter current ratings of 4A, 6A, 10A and 15A. Infineon Technologies www.infineon.com

from Microchip Technology Inc. The single-chip M19122/3 accepts a high-voltage input (up to 42V) while simultaneously regulating a wide output voltage range (0.3V to 16V) without external components or drivers. The internal PIC microcontroller in the M19122/3 can dynamically adjust the operating frequency, overand under-voltage lockout thresholds, current limits, soft-start, voltage or current output set points and maximum duty cycle. Microchip Technology www.microchip.com/M19123

Lighting Smart LED light Opple has unveiled the LED T5 plastic batten. This smart and cost-effective light is Opple’s latest addition to its range of professional and home luminaries.

DC-DC converters The DCM VIA family from Vicor Corporation are ruggedised modular DC-DC converters in a thermally adept, low-profile (9.3mm) VIA package. These offer enhanced functionality and performance, including EMI filtering, transient protection, inrush current limiting, as well as a secondary-referenced control interface for trim, enable and remotesensing. Two new MIL-COTS (M-Grade) versions are available, with 270V nominal input voltage (160-420V input voltage range) and nominal output voltages of 24V and 28V with up to 500 watts of output power. Vicor Corporation www.vicorpower.com

Power analogue buck controller A new Digitally Enhanced Power Analog (DEPA) buck controller for DC-DC power conversion is now available 106

June 2017 | electronics for you

LED T5 comes in a plastic unibody design with an inbuilt integrated connector, so it can be installed anywhere and at any place. Variable length makes it convenient for smaller spaces as well. Priced at ` 395 onwards, this product is available in power capacities ranging from 5W to 20W. Opple Lighting www.opple.com

Security search light BINAY has developed a long-range, solidstate security search light. This product has a narrow beam for enhanced long-range visibility (of one km approx.) According to the company, it is not a standard 500W product, but has specialised optics for a very narrow beam. Consuming 500W max., the light features colour temperature of 5000-6500K, colour rendering index of 75-80, intensity greater than 1,000,000cd with power factor of 0.95 and total har-

monic distortion less than 20 per cent. Binay Opto Electronics Pvt Ltd www.binayLED.com

misCeLLaneous UPS Vertiv, formerly known as Emerson Network Power, has introduced Liebert eXM uninterruptable power supply (UPS) for IT spaces and light industrial applications with medium power requirements. The three-phase Liebert UPS is available in fixed capacities of 80kVA/kW, 100kVA/kW, 120kVA/kW, 160kVA/kW and 200kVA/kW, with an optional builtin isolation transformer. It delivers high efficiency levels for mid-size data centres, with operating efficiencies as high as 97 per cent in double-conversion mode and 99 per cent in energy-saving Eco mode. Vertiv www.Vertivco.com

New system-on-modules and microserver evaluation platform Toradex has partnered with Christmann Informationstechnik + Medien GmbH, , which now offers RECS|Box Atlas Quad Apalis evaluation platform, and RECS|Box Antares and Arneb 1RU server enclosures that can run 24 and 72 Apalis systems-on-modules (SoMs), respectively, in a cluster configuration. Apalis ARM-based SoMs are available in different configurations such as NVIDIA TK1 SoC using a 2.1GHz quad-core Cortex-A15 and a GPU with 192 CUDA cores. The Atlas kit s four Apalis SoMs in an evaluation setting on the ’s desk. The same baseboard can also be used in bigger installations with even more modules. Toradex https://www.toradex.com/ www.efymag.com

First Look estyle

Price ` 17,499

iBall introduces Tablet PC with ‘Remix OS’ Suited for productivity and multi-tasking iBall Slide Brace-X1 4G runs Android based operating system ‘Remix OS’ from Jide Technology, offering you a fusion of mobile and desktop experience. Unlike Android OS devices, which are optimised for consuming content rather than creating/editing it, the new tablet PC s multi-tasking, working on multiple windows, taskbar functionalities that let you do a lot more, keyboard and mouse optimisation, and advanced file management. iBall Slide Brace-X1 4G comes with a 25.6cm (10.1-inch) screen with flexible kickstand. Powered by a 1.3GHz octa-core processor, it packs 2GB RAM and 16GB internal storage that can be expanded up to 64GB via a microSD card. It sports an 8MP rear camera and a 5MP front shooter for selfies. The tablet PC is powered by a 7800mAh battery.

Zopo Speed X A dual-camera smartphone with LED light on the front Chinese smartphone brand ZOPO is expanding its Speed product line-up with the launch of a highend smartphone, the Speed X. The new smartphone comes with a 13MP primary camera, which works with a 2MP secondary camera that allows you to take stunning Price pictures with bokeh efNot announced fect, face beauty mode, yet time lapse mode, and much more. The soft light LED flash on the 13MP front camera can be always on to keep you in the spotlight. It’s like holding a lamp on the front, giving your face a soft radiant complexion. So good-looking selfies can be taken anytime anywhere. The stylish Speed X weighs only 131gm. www.efymag.com

EFY BuREAu

Daiwa launches Toughened Glass LED TV

Price ` 12,999

Perfect for both gaming and home theatre Daiwa has launched its newest LED TV—81cm (32-inch) D32C3GL with toughened glass protection while maintaining clear and vivid quality of high-definition TV. The TV sports HRDP technology for fine details and consistent colour, and A+ grade Samsung display for a wide viewing angle of 178/178 degrees. The advanced Cinema Zoom mode pre-installed in the TV lets you enjoy a movie exactly how the maker wanted it to be. The TV can also play fast-moving games by achieving a low input lag time of just 6.5ms. The TV sports Bluetooth connectivity and powerful audio (10W+10W speakers) are paired with a powerful remote. The remote s not just the TV but also media players and wireless headphones connected to your home entertainment system. The TV comes with two USB and three HDMI ports, along with the convenient on-screen browser and USB ports for easy data transfer from one USB to another. An in-built energy saving function s ultra-low power consumption of less than 38 watts.

Alcatel launches Entry-Level Phablet A 15.2cm (6-inch) phablet with 4G VoLTE connectivity Alcatel has launched a 15.2cm (6-inch) phablet with a comprehensive experience in audio and camera. The new PIXI 4 offers clear, sharp shots for selfies, along with Polaroid filters, a powerful music experience with sound backed by Arkamys, and a super slim design. Continuing its 12.7cm (5-inch) PIXI 4 legacy that had features like Android 5.0 Marshmallow, Quad Core 1.0GHz processor, G sensors and proximity sensor, the new phablet promises to deliver an even better smartphone experience at an affordable price through its technological brilliance. Just 8.3mm thick, it is said to be among the slimmest models in entry-level segment. The phablet comes in metallic silver and metallic gold colours.

Price ` 9100

electronics for you | June 2017

107

first look Price ` 87,999

Acer Gaming Notebook Play your best with powerful audio, video and speed

The Acer Aspire VX 15 is powered by the latest 7th Gen Intel Core i7 processor and NVIDIA GeForce GTX 1050 graphics card. It has an aggressive gaming design with iron red backlit keyboard. Dual fans boost up the thermal system for smooth gameplay, which enhances the experience in every intense battle along with Dolby Audio and Acer TrueHarmony. The notebook comes with 1TB hard drive storage and 128GB M.2 SSD, allowing s to store more multimedia files and enjoy faster speed. All essential ports are available and designed to keep connected cables out of way. The input/ output ports are placed well on the both sides.

Ambrane Bluetooth speaker

HyperX Launches Gaming Headset

A lightweight and compact speaker with playback for up to seven hours

Dolby surround sound at the push of a button

Ambrane India has introduced a compact Bluetooth speaker, BT 8000. The speaker combines high-quality sound with an inte-

Price ` 3199

grated, music-synced light show, bringing the energy and fun of club experience to anyone on the go. With conventional speakers, stereo sound can be heard in only a few ‘sweet spots.’ However, the BT 8000 speaker includes precisely positioned drivers to provide a large, consistent sound field for outdoor listening. Oval shaped, the speaker is claimed to provide the power and sound fidelity of speakers three to five times its size. Powered by a 3600mAh rechargeable battery, it lets s play music for as long as seven hours straight. The inbuilt amplifier delivers minimal distortion with low frequencies. The speaker sports wireless and Bluetooth connectivity.

HyperX, a division of Kingston Technology Company, has introduced its most advanced gaming headset to date, the Cloud Revolver S with plug-and-play Dolby 7.1 surround sound. The Dolby Headphone technology added into the USB dongle delivers high-end immersive audio at the touch of a button. HyperX signature memory foam and a new wide headband design enhance comfort for endless hours of game play. The headset also features multi-platform , 50mm drivers and a noise-cancelling microphone designed for use by professional gamers and eSports teams globally. The Cloud Revolver S sound stage features unique design driver technology that allows gamers to visualise sounds placed around them at different angles and distances. With three preset equaliser settings that separate boost, flat and vocals, they can choose their favourite setting.

Price ` 12,999

GizMo ByTes Facebook Messenger rolls out Instant Games Now gaming fans need not wait for a game to from apps store. Facebook has rolled out no- Instant Games on Facebook Messenger. What’s more, Instant Games now turn-by-turn games instead of just single-player games with score leaderboards. Facebook is also planning to add features like game bots that communicate with players about new levels or rewards, and a dedicated Games tab in Messenger with a curated selection of top games.

Google AutoDraw turns scribbles to art Google’s AutoDraw web-based tool pairs machine learning with drawings created by talented artists to help you draw. Pretty similar to what a handwriting recognition software does, AutoDraw can identify what you wish to draw from a simple scribble and place an appropriate drawing on the screen. It also asks whether you meant something different, similar to what an autocorrect algorithm does on a smartphone keyboard.

‘Piano by Gismart’ adjudged the best music app Piano by Gismart, a music app that lets music lovers play their favourite songs on piano without having any musical skills, has won MediaPost’s 2017 Appy Award in the Best Music App category. The app allows s to play popular songs on piano and create music in a fun way. The gaming aspect makes learning fun. To mix things up, Piano allows s to make the piano sound like an electric guitar or organ, among other musical instruments.

Microsoft announces new features for OneDrive mobile apps Microsoft has added offline folders to OneDrive mobile apps, so s can save folders to their mobile device and open them when they don’t have an Internet connection. “Changes made by other s to the files while you’re offline will automatically be updated when you have an Internet connection again,” wrote Jeff Teper, corporate vice president for the Office, OneDrive and SharePoint teams, in a blog post. The company has also unveiled an iMessage app that lets s easily share documents saved on OneDrive.

The prices, features and specifications are based on information provided to us, or as available on various websites and portals. EFY cannot vouch for their accuracy.

108

June 2017 | electronics for you

www.efymag.com

do-it-yourself estyle

Turn Your Phone Into A Walkie Talkie Jagmeet singh

A

t India Electronics Week 2017 held recently, we ended up using walkie-talkies a lot. Managing two handheld devices—mobile phone and walkie-talkie—proved to be quite a challenge. “Can we get rid of walkie-talkies and use our mobile phones to do the same things?,” asked a colleague. I volunteered to find an answer to that question, and came up with a couple of possibilities. There are apps that can make your mobile phone act like a walkie-talkie. Many of them are free. I reviewed many such (free) apps and here are the best five I found…

Voxer Voxer is a renowned walkie-talkie app for Android and iOS devices that lets you communicate through text, photos and videos alongside live voice . It is touted to be the only walkie-talkie messenger with end-to-end encryption. The app provides you an option to even choose between the handset’s earpiece and loudspeaker. You can start communicating using the Voxer app after enabling an active 3G, 4G or Wi-Fi connection. You can listen and respond to Voxer messages even from a Web browser. Likewise, you can broadcast your messages to up to 500 people at once. A private chat option exchanges messages just between the intended parties.

Zello If you want to create your own private channel to talk with your group privately or a public channel to spread your word, you can pick Zello. The app uses a proprietary lowlatency, push-to-talk (PTT) protocol to enable conversations while on the go. www.efymag.com

integration to let you connect with your Heytell tttalk Bluetooth friends or colleagues walkie-talkie on Facebook and Yes Yes No Twitter. An optional Yes Yes No for voice changer takes the No No Yes voice messaging exNo No Yes perience way beyond usual PTT apps.

comparison of different walkie-talkie apps features

Voxer

Zello

Wi-Fi

Yes

Yes

Mobile data (2G, 3G Yes or 4G)

Yes

Bluetooth

No

No

Easy access without any

No

No

Android

Yes

Yes

iOS

Yes

Yes

Yes

Yes

No

Yes

Yes

Yes

It s channels that can have up to 2500 . Similar to other modern walkietalkie apps, Zello provides you with voice history and real-time call alerts. The app uses Wi-Fi, 2G, 3G and 4G to enable wireless connectivity. The Zello app even enhances the experience by ing Android Wear-based smartwatches and watchOS-powered Apple Watch. However, the additional assistance is limited to push notifications.

HeyTell HeyTell is among the most popular walkie-talkie apps with over a billion s. The app has cross-platform for Android, iOS and Windows phone and is designed to offer instant talk at the press of a single button. The HeyTell app uses an active Internet connectivity on your smartphone to enable two-way communication. It also uses geolocation services to offer international coverage without using traditional PTT technologies such as Bluetooth and Wi-Fi. You need to with a name and on the HeyTell app to begin your walkie-talkie operations. There is also social network

TTtalk

This free app for Android and iOS provides you with instant messaging alongside PTT functionality. You can use emoticons, stickers and audiocons to chat with your team. The app also lets you play any of the previously received voice messages. Unlike HeyTell, the TTtalk app requires a genuine phone number with data connectivity, but it has been optimised for low-bandwidth networks.

Bluetooth Walkie-Talkie Opt for Bluetooth Walkie-Talkie if you do not want to spend your entire data pack on conversations. The app works the moment you pair two devices using Bluetooth. Once paired, you just need to press the on-screen ‘Push to Talk’ button to record your message. Releasing this button transmits the message to the other party. The main drawback of the Bluetooth Walkie-Talkie app is that it’s only available on iOS running iPhone and iPad devices. However, it is the best suited and the easiest walkie-talkie alternative if you own Apple hardware. Jagmeet Singh is assistant editor at EFY electronics for you | June 2017

109

Buyers’ Guide estyle

Wireless HeadpHones Cut The Cord

W Paromik Chakraborty is a technical journalist at EFY

e have already deep-dived into the age of ‘compact’ and ‘hands-free,’ where “less is the new more.” Even headphones manufacturers have found a way to brilliantly improve their products to fit the desires of customers. Apple’s bold step to exclude the conventional 3.5mm headphone jack from its latest iPhone 7 smartphone series for the purpose of space utility maximisation and water-resistance is a case in point. New players too are not far behind. For instance, Seattle-based Human Inc. is coming up with a new product called ‘Sound’ that gestates touch-based earphone control along with language translation and multiple audience connect. Such developments give reasons enough to believe that the future of wireless headphones is now.

How Bluetooth headphones work Wireless headphones or earphones work on digitally transmitted waves—via infrared (IR), radio-frequency (RF) waves or Bluetooth—from source. While each device has its own benefits and drawbacks, in this article we will restrict ourselves to Bluetooth-based headphones as these are portable, flexible and in demand. Bluetooth-driven wireless headphones consist of a receiver that receives the signal transmitted from the Bluetooth device, a small amplifier to strengthen the signal and a battery pack to run the system. The device with Bluetooth (smartphone, tablet, laptop among others) has a small computer chip inside that contains the Bluetooth radio module and software that enable the two devices to pair. Bluetooth digitally transmits the audio through a low-power radio signal, which is received by the receiver of the headphone.

The checklist Wireless headphone designs—in-ear, on-ear and over-ear (from top to bottom)

110

Sound quality. Composed of magnet, voice coils and a diaphragm, the driver unit of a headphone converts electrical energy into sound pressure. Larger-diameter drivers mean

June 2017 | electronics For You

bigger speakers, consequently giving stronger audio output. However, the output quality directly depends on driver components. On an average, driver diameter is 8 to 12mm for earphones and 20 to 50mm for headphones. Sound quality can be judged from factors like sensitivity and sound pressure level (SPL), impedance and total harmonic distortion (THD). SPL decides the loudness range of a headphone. On an average, most headphones have an SPL in the range of 85 to 120dB SPL/mW. Scale of impedance (the resistance to the current that creates the sound) decides the life of the device. Impedance of basic-level headphones tends to be lower (around 32 ohms), requiring less current to work. But this may create background noise. High-end headphones have a higher impedance (up to 330 ohms) and require internal amplifiers for operation, but this also prevents early headphone damage. THD level indicates distortion of sound under high volume. Lower THD is always desirable, and on an average, headphones have less than 1% THD. Design. Select the headphone design as per your intended application, comfort and personal preferences. Headphones come in in-ear, on-ear and over-ear designs. In-ear (canal) design sits in the ear canal directly, providing more impactful sound and better exclusion of external noise. In on-ear (supraaural) design, pieces rest on top of the ear, preventing trapping of heat on the ears. This design is, however, less effective in noise blocking. Over-ear (circumaural) design covers ears completely, ensuring better quality and range of sound. The huge size of the design accommodates a bigger engine, amplifying the sound output (especially the bass experience) by multi-fold. Battery life. Wireless headphones may come with a built-in rechargeable or disposable battery, the former being more common in popular brands. On an average, they use www.eFYmag.com

buyers’ guide a 300mAh lithium-ion polymer battery that provides playback for up to ten hours. High-end products pack battery capacity of around 600mAh, allowing a run-time of 20 hours or more. The built-in battery can be charged by a standard mini or microUSB cable. It takes three to four hours to fully replenish the battery. aptX . aptX is an upgraded codec, presently owned by Qualcomm, which enhances the audio quality via a bitrate efficiency algorithm—providing enhanced sound quality through Bluetooth. The prerequisite is that both the source device and the audio equipment must have aptX to utilise the feature. Presently, most of the popular audio equipment and Android smartphones and devices and have aptX built into them. Android devices without aptX can also be upgraded. MAC and Windows equipment may also be aptX-compatible, depending on their hardware. A higher upgrade, called aptX HD, is also out there, which provides even better audio quality than CD-scale. However, the number of devices ing aptX HD is limited presently. Noise cancellation. Headphones with noise-cancellation capability have built-in microphones and electronic chips that record the ambient noise and create anti-phase sound waves (same amplitude but completely inverse phase) accordingly, which cancel out the original noise. This is very effective against low-frequency noise. High-frequency noise, however, is not controlled much by this technique. Usually, noise-cancellation capability varies within products based on the quality of components. If tested separately, it can be measured by the amount of decibels that is decreased for a given frequency of sound. The performance completely depends on the value of the sound frequency. For example, a headphone that claims to drop 30dB sound may be doing so at 60Hz but might cancel only 10dB at 480Hz, and so on. www.eFYmag.com

Indian market: Latest offerings The Indian market has a substantial range of wireless headphones for buyers to select from. Wireless headphones cost more than their wired counterparts, but provide unparalleled sound experience. High-end products. The latest offering from the stable of Bose, QuietComfort 35, costs around ` 30,000. Featuring the company’s innovative noise-cancellation technology and 20 hours of battery life, these headphones are wrapped in a polyester-based sophisticated covering material, corrosion-resistant stainless steel, glass-filled nylon and impact-resistant materials. Momentum Wireless Over-Ear by Sennheiser costs around ` 35,000. It features NFC and Bluetooth connectivity with aptX , NoiseGuard active noise-cancellation technology and hands-free Voice Max microphone, making it very easy to use. Mid-range products. In the medium price range, Plantronics has introduced models like BackBeat Pro 2, BackBeat Go 3 and BackBeat Fit to suit diverse application areas like sports, work and travel. These headphones cost between ` 6000 and ` 11,000. Another player, Jaybird, Logitechowned sports headphone specialist, has come up with the Jaybird X3 that features ive noise-isolation technology, sweat-proof built with X-Fit System, and in-line remote. It costs between ` 13,000 and ` 19,000. Low-end products. Wireless headphones in this segment are priced between ` 1000 and ` 5000. Companies like Zebronics, Philips, Skullcandy and Sportszee are producing and delivering quality wireless headphones keeping the consumers’ price limitations in mind. You can also find an array of base-range headphones from other players, like LG Tone+ HBS 730, Nokia BH 505 and Jabra Halo Smart. electronics For You | June 2017

111

Do-it-yourself

. S.C eDi iv w D & eo i Th San

Auto ReveRsible DC MotoR With Speed And Rotation Control Ashutosh M. BhAtt

D

C motors are widely used in different machines, mechanisms and motion controls in manufacturing and processing industries. In these machines, it is often required to control speed, direction and the number of rotations of the motor. Some applications require only speed control, some only rotation control and some only direction control. Many applications require control of two or all three Fig. 1: Author’s prototype with motor-sensor arrangement parameters. Some examples are: 1. To release the thread (or Parts List wire) of desired length from a Semiconductors: bobbin, it is required to rotate the IC1 - MOC7811 opto-interrupter IC2 - ATmega16 microcontroller bobbin for a specific number of IC3 - L293D motor driver rotations. Also, there is a need for T1 - 2N2222 npn transistor controlling the speed at which the LED1, LED2 - 5mm LED Resistors (all 1/4-watt, ±5% carbon): thread is released. R1 - 330-ohm 2. In a coil-winding machine, - 1-kilo-ohm R2, R4-R6 it is required to make a specific R3 - 470-ohm R7-R11 - 10-kilo-ohm number of coil turns. This is done R12 - 100-ohm by rotating the motor for a speVR1 - 1-kilo-ohm preset cific number of rotations. Also, it is Capacitors: C1 - 10µF, 16V electrolytic required to vary the motor speed to Miscellaneous: speed up/down the operation. CON1, CON2 - 2-pin terminal connector 3. In a conveyor belt applicaLCD1 - 16×2 LCD module M1 - 12V DC motor tion, to move the object to an exact S1-S4, S6 - Tactile switch distance, it is required to rotate S5 - SPDT switch the motor for a desired number - 5V DC and 12V DC regulated power supply of rotations. - 40-pin IC base 4. In an automatic storage and - Metal strip for attaching on retrieval system, it is required to motor shaft rotate the motor clockwise and anti5. In a robotic arm, to pick and clockwise to move up and down, or left and right. Also, the motor has to place an item, it is required to rotate the motor clockwise and anticlockbe rotated for a desired number of rotations to reach a specific position. wise for a desired number of rota112

June 2017 | electronics For You

tions. Speed control is also required for slow or fast operation, accordingly. There are many more examples and applications where it is required to control speed, direction and rotations of a DC motor. This project is a demonstration of controlling all three parameters. It is an auto-reversible DC motor that automatically reverses when it completes the desired number of rotations. It is an example of a closed-loop control system that utilises loop. It counts the actual number of motor rotations and gives to the system. The system compares a set of rotations with actual motor rotations, and when these match, it reverses the motor. It can repeat this operation in a continuous loop if repeat mode is selected. It also controls the speed of motor from 10 to 100 per cent. The motor starts with 10 per cent speed and gradually attains the selected speed as it completes three to six rotations. Also, when the motor reaches the last six to ten revolutions, the speed gradually decreases, and for the last two rotations it becomes 10 per cent. This speed control is used to precisely stop the motor at the desired number of rotations point and for smooth operation. The project utilises an ATmega16 microcontroller to control the DC motor, an LCD display to show www.eFYmag.com

Do-it-yourself

M1

M1 = 12V DC MOTOR

6 14 11

OUT2 OUT4 OUT3

15

IN4

10

IN3

3

4

5

GND OUT1

GND

GND

GND

IC3 L293D IN2

IN1

R10

S1

S2

14

R8

R9

R7 − R10 = 10K

UP

DOWN

ENTER S3

13

S4

12

11

10

ROTATE

9

8

7

6

5

4

2

15 RS

A/VEE

EN R/W D0 D1 D2 D3 D4 D5 D6 D7

VDD

2

3

7

8

VCC2

16

VCC1

1

EN1

9

EN2

GND CON2 FOR 12V

R12 100E 16 K

LCD1 16x2 LCD VO

Vss

www.eFYmag.com

13 12

CCW

LED1

CW

LED2

R6 1K R5 1K

Fig. 2: Circuit diagram of the auto-reversible DC motor control

XTAL1 XTAL2

GND

24 23 22 31 PD5 PD6 PD7 GND

C1 10u 16V

19 20 21 11

PC2 PC1 PC0

25 PC3 PD4

RST PD0 PD1 PD2 PD3

S6 RESET

4 1

R11 10K

2 3

MOC7811

18

38 37 36 35 34 33 32 29 28 27 26

PA0 PA1 PA2 PA3 PA4 PA5 PA6 PA7 IC2 AREF ATMEGA16 PC7 PC6 PC5 PC4

AVcc Vcc

1 2 3 4 5 6 7 8 9 14 15 16 17

PB0 PB1 PB2 PB3 PB4 PB5 PB6 PB7

10

IC1

R3 470E

T1 2N2222

CON1 FOR 5V

R1 R2 330E 1K

R4 1K

P

2 SINGLE

REPEAT 1 S5

30

40 39

R7

VR1 1K

1

12

The author's prototype, including motor-sensor arrangement, is shown in Fig. 1. 13

various parameters and an optointerrupter sensor to count motor rotations and provide .

Circuit and working Circuit diagram of the auto-reversible DC motor control is shown in Fig. 2. It is built around opto-interrupter module MOC7811(IC1), ATmega16 microcontroller (IC2), 16×2 LCD (LCD1), motor driver L293D (IC3), 12V DC motor (M1) and a few other components. Four push-to-on button/tactile switches (S1 through S4) are connected to port C pins PC0, PC1, PC6 and PC7 of IC2, so that when a button is pressed, it gives a high logic (1) input to the corresponding pin. These switches are used for various settings to control the motor, including up, down, enter and rotation. LCD1 is connected to port A and port D of IC2. Its data pins are connected to port A, and the control pins EN and RS are connected to PD0 and PD1, respectively. LCD1 displays the number of rotations and speed of the DC motor. Port D pin PD7 and port B pin PB3 of IC2 drive the DC motor through motor driver chip L293D. These pins are connected to input pins 2 and 7 of the L293D, and output pins 3 and 6 of the L293D are connected to the motor (M1) terminals. PB3 and PD7 are PWM output pins. PWM output on these pins will vary the speed of motor. Also, when PB3 generates PWM, PD7 is off and the motor rotates in forward (clockwise) direction, and when PD7 generates PWM, PB3 is off and motor rotates in reverse (anticlockwise) direction. Opto-interrupter sensor MOC7811 consists of an IR LED and a phototransistor. The internal IR LED is forward-biased to keep it continuously on. A current-limiting resistor (R1) of 330-ohm is used to limit the current flowing through it. Output of the phototransistor is given as input to the base of transistor T1 that is connected in switch mode. Final output of the sensor circuit is taken from collector of T1. This output is given to external interrupt 0 input pin PD2 of ATmega16. electronics For You | June 2017

113

Do-it-yourself

Fig. 3: Actual size PCB layout of the auto-reversible DC motor control

Fig. 4: Components layout of the PCB

SPDT switch S5 is connected to port B pin PB0, so that when it is connected to +5V it gives logic 1 input, and when connected to ground it gives logic 0 input to pin PB0. S6 is used as the reset switch for IC2. LED1 and LED2 are connected to pins PB6 and PB7 of IC2 for anticlockwise and clockwise indications, respectively.

Circuit operation The motor is initially in stop position. When power is connected to the circuit, “Set rotations” message is displayed on LCD1. Here, you need to enter the required number of motor rotations by pressing S1 or S2. 114

June 2017 | electronics For You

These two switches work as up and down arrow keys for the number of rotations and speed, respectively. S1 is used to increase the number of rotations by five, and S2 is used to decrease the number of rotations by five. The minimum number of rotations is ten. Once the desired number of rotations is selected, press Enter (S3). The LCD display will show “Rotations set

efy Note

The source code of this project is included in this month’s EFY DVD and is also available for free at source. efymag.com

to xx” message. After two seconds, you will be prompted to enter the required speed, when the LCD displays “Set motor speed” message. You can set the speed from 10 to 100 per cent in steps of 10 by pres or down arrow keys. Once the desired speed is set, press Enter. The display now shows “Speed is set to xx%” message. Now, you will be prompted to start motor M1 through “Press rotate” message on LCD1. When you press the rotate button (S4), microcontroller IC2 starts applying PWM on PD7, as per the selected speed, and the motor starts rotating anticlockwise. LED1 turns on to indicate that the motor is rotating in anticlockwise direction. PWM width is increased to the desired speed in two or three steps, so that the motor speed increases step by step. As the motor rotates one revolution, a light thin metal strip attached to the shaft of the motor es through the gap of the optointerrupter sensor module. This interrupts IR light falling on the phototransistor for a fraction of a second. The phototransistor gives a very short-duration positive pulse, which is inverted by transistor T1, and it gives a short-duration negative pulse to pin PD2 of IC2. This negative pulse is ed as one count by IC2. In this way, IC2 counts the number of interrupts as the number of rotations of the motor and displays it on LCD1. As the motor reaches the last six to ten rotations, PWM width is reduced to decrease the speed. For the last two to four rotations of the motor, its speed is reduced to 10 per cent to further slow down the motor. As the motor completes the rotations in anticlockwise direction, PWM is generated from PB3 pin. The motor starts rotating in clockwise direction. LED2 turns on. Again, motor speed is increased in steps to full speed and then decreased in steps to the miniwww.eFYmag.com

Do-it-yourself mum as it completes the desired number of rotations. At this time, if S5 is at position 1 (repeat) and input is high at PB3 pin, the motor will continue to rotate in clockwise and anticlockwise directions at selected speeds. But if S5 is at position 2 (single) and input is low, the motor will stop after one rotation. After two seconds, you will be prompted to enter the desired number of rotations for the next operation, and the cycle will continue.

Software ATmega16 is the main building block of this project because it performs the following tasks: 1. Takes input from pushbutton switches to set speed, rotations, run motor and so on. 2. Rotates the motor at the desired speed using PWM technique. 3. Displays motor speed, rotations and various messages on the LCD. 4. Counts actual motor rotations. 5. Automatically reverses the motor when it completes the desired number of rotations. 6. Rotates the motor clockwise or anticlockwise in repeat mode. To implement all the above functionalities, the software program is loaded into the internal ROM (flash) of the microcontroller. The program (DCMotor_speed.c) is written in C language and is compiled using AVR Simulator 2 available with AVR Studio IDE software. The hex code generated is used to burn into ATmega16 using a suitable programmer. (ProgISP programmer was used at EFY Lab for programming the hex code.)

Construction and testing An actual-size, single-side PCB layout for the autoreversible DC motor control is shown in Fig. 3 and its components layout in Fig. 4. After assembling the circuit, enclose it in a suitable box. Fix CON1 at the back side of the cabinet and S1 through S5 at the top side of the cabinet. Fix S6 on the front side of the cabinet for resetting the circuit. Attach a light thin metal strip to the shaft of motor in such a way that when shaft rotates the metal strip should through the slot (cut) of the MOC7811. After assembling the circuit on the PCB, crosscheck for any wrong connections. Burn the program (DCMotor_speed.hex) into the microcontroller using a suitable programmer. It is recommended to use a 40-pin IC base for ATmega16. Use a 5V DC power supply for the circuit and a 12V DC power supply for the DC motor. Ashutosh M. Bhatt is M.Tech in embedded systems. Currently, he is a lecturer of electronics and radio engineering at Government Polytechnic, Jamnagar, Gujarat

www.eFYmag.com

electronics For You | June 2017

115

Do-it-yourself

hi Nid ria u h Kat

Ultra-low-Power SenSor HUb Using nRF24L01 Modules Somnath Bera

P

the receiver unit is a 2.4GHz transceiver suitable for ultra-low-power wireless applications. It is designed for operation in ISM band (industrial, scientific and medical radio band) of 2.400-2.4835GHz. The nRF24L01+PA and LNA module used in the transmitter unit is similar to nRF24L01+ but has built-in power amplifier (PA) and low-noise amplifier (LNA) circuits, which help transmit the signal to a longer distance and achieve more stable performance for industrystandard applications. The search for a low-power system started when some EFY readers questioned a DIY of mine, saying that ATmega328P processor would not sustain 3.3/5 volts. But it Fig. 1: nRF24L01+PA LNA module worked well because of the extreme low power capability of Different Channels for 2.4Ghz BanD the ATmega processor. More detail is provided Channel frequency (Mhz) Description under extreme low 0 to 82 2400 to 2482 Legal but noisy (conflicts with power sub-head of Wi-Fi LAN, Bluetooth, etc) this article.

resented here is an ATmega328P microcontroller-based ultralow-power sensor hub using nRF24L01+ radio frequency (RF) transceiver modules having transmitter and receiver units. The transmitter sends humidity and temperature values of a place to the receiver kept at a remote location for display on an OLED module. The nRF24L01+ module used in

83 to 99

2483 to 2499

Not legal

100

2500

Licensed channel

Circuit and working

101 to 119 2501 to 2519

Legal, clear and above the Wi-Fi LAN

120 to 125 2520 to 2525

Used in defense

ATmega328P: Active Supply Current vs Frequency (1-20MHz) 5.5V

12

5.0V

10

4.5V

ICC (mA)

14

8

4.0V

6

3.3V

4

2.7V

2 0

1.8V 0

2

4

6

8 10 12 14 Frequency (MHz)

Fig. 2: Supply current vs frequency

116

June 2017 | electronics For You

16

18

20

The heart of the sensor hub is the low-power ATmega328P, which runs at internal 4MHz speed. A DHT22 sensor is used to pick up the relative humidity and temperature of the area. Low-power radio transceiver module nRF24L01+PA LNA (Fig. 1) transmitter transmits the signal. You can add more sensors as lots of GPIO pins of the transceiver

module are unused. A 470µF capacitor acts as the stabiliser, while the sensor (radio) module draws power on burst. Parameters for radios in both the transmitter and receiver modules are set as follows: = 250kBps Data Rate RF_CH = 0x6c [108] = nRF24L01+ Model CRC Length = 16 bits = PA_MAX PA Power RF_CH = 0x6c is the 108th channel. Normally, the 2.4GHz band is divided into a variety of channels as shown in the table. The 108th channel is at around 2.508GHz, which is well above Wi-Fi LAN frequency and legal to use. You can set the module to operate on 250kBps data rate and at different channels. Chances of dissemination are fairly high at 250kBps data rate. Interestingly, almost all microwave ovens operate at 2.4GHz. Therefore channels 0-82 are very noisy. The script for this project is designed such that first all the radio details are printed on the serial terminal and then the data starts appearing. Principle of operation. The ATmega328P processor is a low-power version of ATmega328 chip. Its supply current vs frequency graph, as per the datasheet, is shown in Fig. 2. As can be seen, the processor is quite tolerant from 1.8 volts up to 5.5 volts at 1MHz, 1.8V, 25ºC. The processor works in active mode at 0.2µA, power-down mode at 0.1µA and power-save mode at 0.75µA (including 32kHz RTC). Getting a 0.2µA current for deepsleep power-down mode, however, www.eFYmag.com

Do-it-yourself CON1 FOR NRF24L01+ PA LNA IRQ MOSI CSN VCC

NC GND

R2 10K

1N4007 R1 10K BATT.1 3V − 3.7V

20 21

PD0/RXD

PC5/SCL

28

3

PD1/TXD

PC4/SDA

27

4

PD2

PC3

26

5

PD3

IC1

ATMEGA328PPC2 25

PD4

PC1

24

9

PB6/XTAL1

PC0

23

10

PB7/XTAL2

PB5/SCK

19

PD5

PB4/MISO

18

PD6

PB3/MOSI

17

PD7

PB2

16

PB0

PB1

15

AGND

22

12

C3 470u 16V

Vcc PC6/RESET

6

11

D1 GND

AVcc AVREF

2

CON2 FOR DHT22

VDD

7 1

DATA

C2 10u,16V S1 POWER

C1 470u 16V

S2 RESET

MISO SCK CE GND

13 14 8

GND

Fig. 3: Transmitter circuit CON3 FOR NRF24L01+ IRQ MOSI CSN VCC

C4 470u 16V

MISO SCK CE GND

S3 POWER

7

AVcc

20

AVREF

21

PD0/RXD

PC5/SCL

28

3

PD1/TXD

PC4/SDA

27

4

PD2

PC3

26

1 2

C5 10u,16V

D2 1N4007 GND

S4 RESET

R3 10K GND VCC SCL SDA

CON4 FOR OLED BATT.2 3V − 3.7V

5

Vcc PC6/RESET

IC2

PD3

ATMEGA328P PC2 25

6

PD4

PC1

24

9

PB6/XTAL1

PC0

23

10

PB7/XTAL2

PB5/SCK

19

PD5

PB4/MISO

18

PD6

PB3/MOSI

17

PD7

PB2

16

PB0

PB1

15

AGND

22

11 12 13 14 8

GND

Fig. 4: Receiver circuit

is tricky. At the same time, getting around 1µA in deep-sleep mode is easy, with the help of the chip’s watchdog timer (WDT) and brownout detection fuse. First, turn off the brown-out detection fuse so that low-voltage operation becomes possible without resetting. As the voltage goes down, so does the current (refer the graph in Fig. 2). Interestingly, as the frequency goes down, so does the power consumption. That means if the chip is designed to work on a lower resonating frequency, the power consumption will further reduce. To reduce the frequency of opera118

June 2017 | electronics For You

tion, we divide the 8MHz frequency internally by commands. At 3.3V operation, power consumption goes down as follows: clock_div_1-3.1 mA clock_div_2 - 1.8 mA clock_div_4-1.1 mA clock_div_8 - 750 μA clock_div_16-550 μA cock_div_32 - 393 μA clock_div_64-351 μA clock_div_128 - 296 μA clock_div_256 - 288 μA

To reduce the frequency of operation, here’s a very simple command to use inside setup( ): // slow clock to divide by 256 clock_prescale_set (clock_div_256);

As per Fig. 2, at 4MHz, the chip will continue to work off 1.8V supply. If the WDT is turned off, the chip will get sleep current to the tune

of 1µA, but it will not wake up on its own—unless you give it a shock through interrupter (0) from pin 4 of ATmega328P, momentarily making it ground. However, in this project we want it to send signals periodically (once every 42 minutes). Therefore we don’t set the WDT ‘off.’ The sleep current will be more than 30µA but the operation will be periodic on its own. The WDT clock is not very precise as compared to other methods of time keeping. However, at low power, the WDT clock does not differ much from real-time clocks. The entire operation is accomplished with the help of a library file called lowpower.h, which can be ed from the following link: http://www.rocketscream.com/ blog/2011/07/04/lightweight-lowpower-arduino-library/

Transmitter unit Fig. 3 shows the transmitter circuit. It can be powered by a 3.7V Li-ion cell or two 1.5V pencil cells. Theoretically, the devices used here are capable of running at up to 1.8 volts, but at less than 2 volts the nRF24L01+PA&LNA radio stops working and at less than 2.8 volts the DHT22 sensor will not function. Small capacitors are added to the supply bus of the DHT22 and the nRF24L01 for stability. Vcc supply for the DHT22 is taken from port PD4 (pin 6) of ATmega328P, whereas the nRF24L01+ is directly connected with 3.7V supply. To reduce power consumption by the radio in idle state, we use powerDown( ) command of the radio in the code. Before burning the code (nrf24l01_tx6.ino) into a fresh ATmega328P chip, burn the bootloader code for the inbuilt internal 8MHz clock. For details, see under AVR programmer sub-head.

Receiver unit Fig. 4 shows the receiver circuit. The nRF24L01+ radio is designed www.eFYmag.com

Do-it-yourself to operate between 1.9-3.6 volts only. Connecting the module to 5 volts may damage it permanently; it overheats in case of reverse connection or high-voltage connection. Connect a small capacitor (say, 10µF) between Vcc and GND leads of nRF24L01+ radio. Pin details of nRF24L01+ (shown in Fig. 5) and nRF24L01+PA and LNA modules are the same. The nRF24L01+ module is connected to an I2C 64x128 OLED through an ATmega328P. The humidity and temperature data received by the module is displayed on the OLED. You can also see this data on the serial monitor of Arduino IDE. All the nRF24L01+ radios are capable of transmitting six channels simultaneously. Unique pipe addresses or IDs selected here for both the transmitter and receiver modules are 0xE8E8F0F0E1LL.

AVR programmer To make this project work, you need an AVR programmer. You can make

one yourself as described under ‘Arduino as AVR Programmer’ at http://electronicsforu.com/ electronics-projects/arduino-avrprogrammer This small setup requires an Arduino board, a ZIF (zero-insertion force) socket, a blank ATmega328P and a few ive components. With these, you can create as much Arduino as you want on the fly. First, burn the bootloader for the internal 8MHz oscillator of the ATmega328P chip. To do this, in any latest Arduino IDE, select Tools→Boards→Arduino LilyPad and then burn the Arduino bootloader onto the ATmega328P chip. LilyPad is the simplest Arduino board on internal 8MHz clock. In case the chip is already burnt as Arduino Uno (16MHz clock), it will not burn the 8MHz internal clock bootloader, unless the 16MHz quartz is connected to it. Once the internal 8MHz clock bootloader is burnt, remove the 16MHz quartz and proceed to burn the main sketch.

Extreme low power VCC

CSN

GND CE SCK MISO IRQ

MOSI

Fig. 5: Pin details of nRF24L01+ module

Fig. 6: Actual-size PCB layout of the transmitter

120

June 2017 | electronics For You

A small lithium-ion button cell is good enough to run the small sensor hub and transmit signals up to 500 metres away for more than 25 days non-stop. On a 3.7V cell, the transmitter takes just 6-7mA current for two seconds and then sleeps for 40 seconds drawing 30µA current. The battery runs for weeks together.

The receiver draws 32mA current for ten seconds and then sleeps for 32 seconds on 8mA current. Here’s the calculation: Current consumption (transmitt er)=(2×7)+(40×0.030)=15.2mA seconds in 42 seconds Therefore, in 1 second = 15.2/ 42 mA seconds=0.3619mA seconds In 1 minute = 0.3619×60 In 1 hour = 0.3619×60×60 mA seconds In 24 hours=0.3619×60×60× 24 mA seconds =(0.3619×60×60×24)/3600 mA hours =8.68571mAh On a 3.7-volt, 150mAh battery it Parts List Semiconductors: IC1, IC2 - ATmega328P microcontroller - nRF24L01+PA LNA 2.4GHz transceiver module - nRF24L01+ 2.4GHz transceiver module - DHT22/AM2302 temperature and relative humidity sensor module - 4-pin 2.44cm I2C OLED module D1, D2 - 1N4007 rectifier diode Resistors (all 1/4-watt, ±5% carbon): R1-R3 - 10-kilo-ohm Capacitors: - 470μF, 16V electrolytic C1, C3, C4 C2, C5 - 10μF, 16V electrolytic Miscellaneous: CON1, CON3 - 8-pin berg strip female connector CON2, CON4 - 4-pin berg strip female connector S1, S3 - On/off toggle switch S2, S4 - Tactile switch BATT.1, BATT.2 - 3V-3.7V Li-ion battery - 28-pin IC socket

Fig. 7: Components layout for the PCB in Fig. 6 www.eFYmag.com

Do-it-yourself the voltage drops by 0.7 volt and the radios start working again.

Construction and testing

Fig. 8: Actual-size PCB layout of the receiver

Fig. 9: Components layout for the PCB in Fig. 8

will last for 150/8.68571 = 17 days Current consumption (receiver) is {(10×32)+(32×8.0)}/42 = 13.714mA seconds =(13.714×60×60×24)/3600 =329.143mAh On a 3.7-volt, 150mAh button cell, it will last for about 150/329= 27 minutes. Display consumes more current. Note that nRF24L01 radios operate off at maximum 3.6 volts. Sometimes, when the Li-ion battery is fully charged, the voltage may go as high as 4.01 volts, so the radio may not work. Put a diode in series with the battery connection so that

efy Note

The source code of this project is included in this month’s EFY DVD and is also available for free at source. www.eFYmag.com efymag.com

An actual-size PCB layout for the transmitter circuit is shown in Fig. 6 and its components layout in Fig. 7. PCB layout for the receiver circuit is shown in Fig. 8 and its components layout in Fig. 9. After assembling the circuits on respective PCBs, enclose these in suitable cabinets and place them some distance apart. Connect the battery to the transmitter unit and measure the current. The current withdrawn is 6-7 mA while transmitting for two seconds and 30µA for the remaining 40 seconds. For periodicity, we used the WDT of the MCU, which is not very precise. It provides five incremental steps of 250ms, 500ms, 2, 4 and 8 seconds. That means, it will transmit once every 5×8+2=42 seconds. The DHT22 sensor requires a minimum of 2.8 volts to operate. Therefore put a fairly big capacitor (100-470µF) across its Vcc and GND pins. Else, you will get temperature and relative humidity values as zero. The sensor takes two seconds between readings. Somnath Bera is an avid of open source software. Professionally, he is a thermal power expert and works as additional general manager at NTPC Ltd

electronics For You | June 2017

121

Do-it-yourself

. S.C Di e v i Dw

3W/6W Audio Amplifier Using TDA2003 Raj K. GoRKhali

T

his is a simple audio-frequency (AF) amplifier using the popular audio amplifier IC TDA2003. The IC comes in a 5-pin TO-220B package.

Circuit and working Fig. 1 shows the circuit diagram of the 3W/6W AF amplifier built around IC TDA2003 (IC1), an 8-ohm, 6-watt speaker (LS1) and a few other components. The amplifier IC delivers 3W output power using a 6V, 500mA power supply, and 6W output power using a 12V, 500mA power supply, with an 8-ohm, 6-watt speaker. Pin 1 of TDA2003 is the input terminal, which is connected to ground through a 100-kilo-ohm resistor. The audio frequency (or audio signal) is fed to pin 1 of IC1 through the combination of a 10-kilo-ohm potmeter (VR1) and capacitors C1 and C2. Potmeter VR1 is used as volume control. Pin 3 of IC1 is connected to ground. Pin 4 is the output terminal, Parts List Semiconductors: IC1 - TDA2003 audio amplifier Resistors (all 1/4-watt, ±5% carbon): R1 - 100-kilo-ohm R2 - 220-ohm - 2.2-ohm R3 R4 - 1-ohm VR1 - 10-kilo-ohm potmeter Capacitors: - 2.2µF, 16V electrolytic C1 - 0.1µF ceramic disk C2, C4 C3 - 220µF, 16V electrolytic C5 - 470µF, 16V electrolytic Miscellaneous: CON1 - 2-pin connector LS1 - 8-ohm, 6-watt loudspeaker S1 - On/off switch Batt.1 - 6V battery - 2-pin terminal connector for battery - Heat-sink for TDA2003 - Audio signal source - 6V/12V, 500mA DC power supply (optional)

122

June 2017 | electronics For You

C1 2.2u,16V

C2 0.1u

VR1 10K VOLUME CONTROL CON1 FOR AUDIO INPUT

S1 ON/OFF 5

2

IC1 TDA2003 3

R1 100K

C5 470u,16V

1

C3 220u 16V

4

R2 220E

C4 0.1u

R4 1E

LS1 8−OHM 6W

BATT.1 6V

6V battery or a 6V DC adaptor. Use of a suitable heatsink is recommended for IC1.

Construction and testing

An actual-size, single-side PCB for 3W/6W audio GND amplifier usFig. 1: Circuit diagram of 3W to 6W audio amplifier using TDA2003 ing TDA2003 is shown in Fig. 2 and its components layout in Fig. 3. After assembling the circuit on the PCB, enclose it in a suitable box. Solder TDA2003 and other components using a 25W soldering iron. Use a 2-pin connector for input and output connectors, to make your prototype safe and clean. Fig. 2: Actual-size PCB layout of 3W to 6W audio For testing the circuit, amplifier using TDA2003 connect a 6V battery to the circuit. Also, connect an 8-ohm, 6-watt speaker to LS1. If a 12V DC supply is used, voltage ratings of C1, C3 and C5 should be 25V or above. Take a metal screwdriver and gently touch at input pin 1 of IC1. If your circuit is wired properly, you will hear Fig. 3: Components layout for the PCB a humming sound from the speaker. Else, vary potmeter VR1 to increase the volume until humwhich is connected to one terminal ming sound is heard from the speaker. of the speaker through a 470µF, 16V Now your circuit is ready to use. capacitor (C5). It is also connected to ground through a 0.1µF capacitor (C4) and a 1-ohm resistor (R4). Raj K. Gorkhali is a regular contributor to EFY and has Pin 5 of IC1 is connected to +6V a keen interest in deg power supply via switch S1. Power electronic circuits supply can be provided through a R3 2.2E

www.eFYmag.com

Do-it-yourself

S Dw .C. ive Di

Dual-Channel, Active Ac AnAlogue Probe Petre tzv Petrov

P

resented here is a low-cost, dual-channel, active AC analogue probe. It is suitable for oscilloscopes, multimeters and other analogue measurement equipment. When experimenting with AM receivers for long, medium and short waves, s need a highimpedance analogue probe with input capacitance below 20pF. The required input impedance from the probe at low frequency is usually one mega-ohm or above. Most of the low-cost ive probes for oscilloscopes have input capacitance well above 100pF. If you include the input capacitance of the oscilloscope, the total capacitance becomes even higher. The bandwidths of these low-cost ive probes are usually limited to 1-2MHz. These cannot be connected in parallel with L-C networks in AM receivers because that will change the parameters of these networks significantly.

Circuit and working Fig. 1 shows the circuit diagram of the dual-channel, active AC analogue probe. It is built around two n-channel PN4393 JFET transistors (T1 and T2) and a few resistors and capacitors. Most of the oscilloscopes come with two channels, so you need a dual-channel active probe with common ground and common power supply. At low frequency, the input resistance of the probe, determined mainly by resistors R3 and R4, is around 10-mega-ohm. At high frequency, the input impedance is determined by the total input cawww.efymag.com

VR1

C3 100u

R5 10K

10K

25V

VR2 10K

R10 220E

G

T1 PN4393 R4 10M

T2 S

PN4393 S

R11 1M

R1 1K

R6 470E

R2 1K

CON2 FOR INPUT2

C2 0.33u

C8 0.33u

R8 470E

C9 0.1u

C10 0.33u J3

R7 470E

J4

CON3 FOR OUTPUT1

R12 220E

C7 0.1u

C1 CON1 0.33u FOR INPUT1

25V

D

G

R3 10M

CON5 FOR 9V

GND

D J1 J2

C6 220u

C5 0.33u

C4 0.1u

R9 470E

R13 1M

CON4 FOR OUTPUT2

Fig. 1: Circuit diagram of dual-channel, active AC analogue probe

pacitance of the probe. With careful implementation and use of appropriate JFETs, you can obtain input capacitance below 20pF. Transistors T1 and T2 should have low noise level and low input capacitance. So you can choose from transistors like PN4391, PN4392 and PN4393 as these have input capacitance of typically 14pF. You can also use transistors J201, J202, J203, 2N5457, 2N5458 and BF245 (with 1.5dB noise figure). High-frequency JFETs in small cases are preferred but these are costly. The voltage between the drain and the source of T1 and T2 is adjusted to around half of the power supply, using resistors R6 through R9 and jumpers J3 and J4. The circuitry using resistor R5, presets VR1 and VR2, and capacitor C3 is used to adjust the offset voltage applied

Parts List Semiconductors: - PN4393 JFET T1, T2 Resistors (all 1/4-watt, ±5% carbon): R1, R2 - 1-kilo-ohm R3, R4 - 10-mega-ohm - 10-kilo-ohm R5 R6-R9 - 470-ohm R10, R12 - 220-ohm R11, R13 - 1-mega-ohm VR1, VR2 - 10-kilo-ohm preset Capacitors: C1, C2, C5, C8, C10 - 0.33µF ceramic disk C3 - 100µF, 25V electrolytic C4, C7, C9 - 0.1µF ceramic disk C6 - 220µF, 25V electrolytic Miscellaneous: CON1-CON4 - 2-pin connector CON5 - 2-pin terminal connector J1-J4 - Shorting jumper - 9V DC power supply/ 9V battery

to the gates of the JFETs. These components are not mandatory. If you choose not to use these, close electronics for you | June 2017

123

Do-it-yourself

Fig. 2: Actual-size PCB layout of dual-channel active AC analogue probe

Fig. 3: Components layout for the PCB

124

June 2017 | electronics for you

jumpers J1 and J2, and connect resistors R3 and R4 to ground. The gain of each of the channels is slightly below unity, but that is not important in most of the cases. Most oscilloscopes produce a square wave reference signal, which can be used for measurement of the gain and evaluation of the probe.

Construction and testing An actual-size PCB layout for dual-channel active AC analogue probe is shown in Fig. 2 and its components layout in Fig. 3. After assembling the circuit on the PCB, enclose it in a suitable cabinet with connectors CON1 and CON2 affixed on the front side and connectors CON3 and CON4 on the rear side of the cabinet.

The probe can be implemented on a small PCB. During testing and adjustment, the inputs of the probe can be connected or soldered to L-C networks using very short (3-4cm long) conductors. Power the circuit using a dry battery or rechargeable battery. The circuit does not require any special adjustment, so you can start using it straightaway after assembly. If you use jumpers J3 and J4, adjust VR1 and VR2 for the best performance of the probe. After proper implementation, the probe can have bandwidths above 10MHz with a signal generator having low output impedance. Petre Tzv Petrov was a researcher and assistant professor in Technical University of Sofia (Bulgaria) and expert-lecturer in OFPPT(Casablance), Kingdom of Morocco. Now he is working as an electronics engineer in the private sector in Bulgaria

www.efymag.com

Do-it-yourself

S Dw .C. ive Di

Electronic Tone GeneraTor SySTem Pamarthi KanaKraja

I

Tone key switches (S1 through S7) are connected using a resistor ladder. The resistors (R1 through R6) are placed in a sequential order, connecting each switch to 5V power supply. Resistor values (in ohms) used are 100k, 10k, 4.7k, 1k, 470-ohm and 220-ohm. S1 is directly connected to 5V power supply. One terminal of switch S2 is connected to 5V supply through R1 and the other terminal to pin A0 of Board1. Switches S3 through S7 are also connected to analogue pin A0 of Arduino Uno board in a similar manner. The Arduino program (tone_ generator.ino) defines the musical notes and frequencies associated with Circuit and working each tone key. Change the frequency values or add additional switches to Circuit diagram of the electronic tone generator is shown in Fig. 2. It is based fully customise your project. LCD screen. Connect the LCD1 on Arduino Uno board (BOARD1), liquid crystal display (LCD1), transistor pins according to the circuit diagram BC547 (T1), a buzzer (PZ1) and a few shown in Fig. 2. Resistor R11 conFig. 1: Author’s prototype nected at pin 3 of LCD1 is used as the other components. contrast control of the screen. Pin 15 of LCD1 is connected to 5V power supR11 ply and pin 16 is connected to ground 1K Vss 1 K 16 LCD1 PZ1 for backlight on the LCD screen. LCD1 VO 3 16x2 LINE LCD PIEZO is configured in 4-bit mode. Its pins A/VEE VDD BUZZER 2 15 D7 D6 D5 D4 D3 D2 D1 D0 EN R/W RS D4, D5, D6 and D7, select (RS) CON1 14 13 12 11 10 9 8 7 6 5 4 FOR and enable (EN) are connected to pins R10 9V S8 10K 5, 4, 3, 2, 12 and 11, respectively, of T1 R7 BC547 the Arduino Uno R3 board. Read/write 10K (R/W) pin is connected to ground. S1 R8 LED1 100E Connect one terminal of menu S2 R1 220E switch S8 to 5V and the other terminal to ground through R10. Also, conARDUINO DIGITAL S3 R2 470E nect it to pin 7 of the Arduino. ATMEGA328 S4 R3 1K Glowing of LED1 indicates that you ANALOG are in Menu mode and the tone keys 9V DC S5 R4 4.7K have been disabled. LED1 turns on S6 R5 10K once S8 is pressed. Connect the anode BOARD1 ARDUINO UNO (positive) of LED1 to pin 6 of Arduino S7 R6 100K Uno and cathode to ground through R8. R9 10K To integrate the piezo buzzer (PZ1) through transistor T1, connect GND the base of the transistor to Arduino’s pin 8 through R7. Connect piezo Fig. 2: Circuit diagram of the tone generator system 7 6 5 4 3 2 TX 1 RX 0 A0 A1 A2 A3 A4 A5

DRIVER

IOREF RST 3.3V 5V GND GND Vin

USB

POWER INPUT

www.efymag.com

GND 13 12 11 10 9 8

tiple tones with the help of Arduino software programming. You can see how the device works, including the menu of preset musical tones, in the author’s prototype shown in Fig. 1. There are five musical tones already programmed in this project. Technically, tones four and five in this project are the same, which are left to readers to create better ones. Writing a musical code was not the goal of the project. Basic coding for main loops and individual tone loops is easy to adapt to your own ideas, so feel free to change the code as per your requirement.

AREF

t would be cool to build an electronic tone generator system yourself. Described here is a small tone generator based on pulsewidth modulation (PWM) concept in which a piezo buzzer generates mul-

electronics for you | June 2017

125

efy Note

Do-it-yourself Parts List Semiconductors: - Arduino Uno R3 BOARD1 - BC547 npn transistor T1 - 5mm LED LED1 Resistors (all 1/4-watt, ±5% carbon): R1 - 220-ohm - 470-ohm R2 - 1-kilo-ohm R3, R11 - 4.7-kilo-ohm R4 R5, R7, R9, R10 - 10-kilo-ohm R6 - 100-kilo-ohm - 100-ohm R8 Miscellaneous: - 2-pin terminal connector CON1 - Piezo buzzer PZ1 - Tactile switch S1-S8 LCD1 - 16×2 LCD module - 9V battery - 9V DC adaptor

sical tones: Super Mario Bros Theme song, Super Mario Bros Underworld song, Jeopardy Theme song, and the fourth and fifth are same modified version of Jeopardy Theme song generated for you to modify. To return to tone keys, cycle through the remaining tones in the menu. Once LCD1 screen is blank and LED1 is off, press each key (S1 to S7) to generate unique beep tones.

buzzer’s negative terminal to the collector of T1 and positive terminal to 9V battery.

Software Circuit operation is done using the software program (tone_generator. ino) loaded into the internal memory of Arduino Uno R3. The program implements all required functionalities including handling inputs with the help of switching interface. The program is written in Arduino programming language. Arduino IDE 1.6.4 is used to compile and the program (called sketch). The sketch begins by importing LiquidCrystal.h and MyTones.h libraries, so you can reference various items from these later on in the code. Next, the code starts by defining a name for S8, inputs versus outputs, and setting the frequency values for each of the seven pushbutton keys (switches S1 through S7). A new tab is created for each individual tone in an effort to keep the code organised and easier to understand. If you wish to delete or add more tones, make sure to change the code in the main program/sketch as well. Main loop. The program begins with the LED1 turned off, and all tone keys (S1-S7) active. Once the loop detects that S8 has been pressed, it disables S1 through S7, and turns the LED1 on to indicate that you are in Menu mode. Accessing the menu. Press Menu 126

June 2017 | electronics for you

The source code of this project is included in this month’s EFY DVD and is also available for free at source. efymag.com

Construction and testing Fig. 3: Actual-size PCB layout of tone generator system

An actual-size, single-side PCB for the electronic tone generator is shown in Fig. 3 and its components layout in Fig. 4. After assembling the circuit, enclose it in a suitable box. Fix CON1, LED1 and S1 through S8 on the front of the box. Connect the 9V using a 9V adaptor to Arduino Uno board and 9V battery for PZ1.

Testing procedure

Fig. 4: Components layout of the PCB

button to cycle through the present tones, and press any key (S1 through S7) to start playing the tone displayed on the LCD1 screen. The LCD1 screen will show the tone that is currently playing and then return to the same point in the menu once the tone has finished playing. From there, you can either replay or continue to play next tone available in the list. Double-click S8 to restart the menu from the beginning. Here, we have created five mu-

After assembling the circuit, the program into Arduino Uno board. Connect 9V adaptor to Arduino Uno. LCD1 is blank and LED1 remains off. Tone switches produce individual tones when each switch is pressed momentarily. Press S8 momentarily and tone switches will get disabled. When you press S8 again, LCD1 shows the message: “Which song would you like to play?” When you press any tone key, the first tone starts playing. On further pressing S8, the next tone will be displayed on the LCD1. When you press any tone key again, the second tone starts playing. Thereafter, repeat the same process till the last tone is played. Press S8 again and the LCD1 will be blank. Pamarthi Kanakaraja is assistant professor at Usha Rama College of Engineering and Technology, Andhra Pradesh

www.efymag.com

Do-it-yourself

Sa Th ni eo

PC-Based OsCillOsCOPe Using Arduino

www.efymag.com

7 6 5 4 3 2 TX 1 RX 0

DIGITAL

ATMEGA328

ANALOG A0 A1 A2 A3 A4 A5

AREF DRIVER

USB POWER INPUT

TO USB

ARDUINO

IOREF RST 3.3V 5V GND GND Vin

O

scilloscopes are an essential tool for electronics hobbyists and professionals to that their designs would work as expected. PC-based oscilloscopes score over standalone oscilloscopes due to their compact size, low cost and ability to do offline analysis. Here we describe how you can make your own oscilloscope at a very low cost using your PC and an Arduino board as the hardware for signal acquisition. You can use this oscilloscope to capture frequency signals up to 5kHz. The Arduino board, the heart of the oscilloscope, reads the values from its inbuilt analogueto-digital converter (ADC) and pushes these to the PC via USB port. We have provided here an Arduino sketch, which you can compile and load directly to the Arduino. You also need to install an executable file or application in your Windows PC. This application works as the front-end to plot input signals as waveforms on your computer screen. The Arduino board consists of Atmel’s AVR microcontroller, which can be 8-, 16- or 32-bit based on the type of the board. For this project you can use any variant of the Arduino as hardware. The AVR microcontroller has an inbuilt ADC. In the project we use pin A0 to capture the input signal. The captured input signal is fed to UART via UART-USB converter in the Arduino to the PC. A virtual COM port is created by Windows whenever the Arduino connects to the PC. A Windows-based application developed using NI LabWindows opens up the virtual COM port and starts plotting signals visually using Graph libraries. The sampling speed of the oscil-

GND 13 12 11 10 9 8

Ramalingam Balaji

BOARD1 LAPTOP D1 1N4148

R1 10K

D2 1N4148

CON1 INPUT

Fig. 1: Circuit of the PC-based oscilloscope using Arduino

Fig. 2: Message on the screen when the PC-based scope is run for the first time

loscope is limited by the baud rate of the UART. The Arduino sketch is coded to read the ADC using ISR, and the UART baud rate is configured at 115200, which sends data at 85µs intervals. This gives an effective sampling rate of 12kSa/s.

Construction The PC scope set-up is quite simple and straightforward as shown in

Fig. 1. The Arduino board connects to your laptop or PC via the USB cable. Any external power supply for the board is not required as the board is powered by the USB only. Connect switching diodes (D1 and D2) as input protection circuit to pin A0 of the Arduino’s ADC. You need Arduino sketch (pcscope.ino) and PC software or executable file (PCScope.exe) in order to use this electronics for you | June 2017

127

Do-it-yourself circuit. Install PCScope. exe program (developed by author) in your Windows PC and open the application. Next, open the Arduino sketch from Arduino IDE and compile the sketch. Connect the Arduino board to the PC and flash the sketch into the microcontroller on the Arduino board. Fig. 3: Message after the hardware successfully connects to the PC The ADC of Arduino can measure voltages up to 5V. So it is advisable to add a small protection circuit to limit the input voltage to 5V and clamp the negative voltage. A low-power, fast-switching diode like 1N4148 can be used to protect the input pin. Connect a 10-kiloohm resistor in series with the input. It will work as a current limiter in case the input goes Fig. 4: Test signal of 525Hz square waveform captured on the screen beyond 5V. Additional voltage dividers can be used in case you need to measure voltages higher than 5V.

which is much faster than the UART data transfer rate. PC software. As stated earlier, the frontend PC software for signal acquisition and processing is developed using NI LabWindows. The serial port data is captured through Arduino at regular time intervals and plotted as a graph on the screen using the Plot function library. The display points along Xaxis are calculated based on the -defined time scale. The Y-axis range is set using the voltage selection control.

Testing

After installing the PC scope application, click ‘Connect’ button on your PC screen to connect to the Arduino board (Fig. 2). When the board gets connected to your PC, you will get a confirmation message for three seconds as shown in Fig. 3. Software Feed any squarewave input of up to 5kHz at Arduino sketch. The CON1. The software must sampling rate of this PC plot its output waveform scope application is limon your PC. Square and ited by the rate at which triangular output wavethe data is sent to the PC. forms of 525Hz and Baud rate of 115000 gives 530Hz captured on the time interval of around screen during testing are 85 µs. It is important to shown in Figs 4 and 5, get the ADC signals much Fig. 5: Test signal of 530Hz triangular waveform captured on the screen respectively. Similarly, before this time to get reyou can feed rectangular or pulse the prescaler to 16. With this, you liable data plotting. The sketch reads inputs (but not sine waves) to get get ADC conversion every 20µs, pin A0 of Board1 and sends to UART output waveforms. at 115200 baud rate. At this speed, bytes of the input are pushed at time intervals of around 85µs. The author is a program efy Note By default, the ADC configuration manager at Robert Bosch, The source code of Bengaluru. He has filed several of the Arduino gives samples every this project is patents in automotive electronics 116µs. So here the ADC is configured and published papers in SAE included in this conferences and several with additional lines of code to get month’s EFY DVD and international magazines samples faster than 85µs by setting is also available for free at source. efymag.com www.efymag.com 128 June 2017 | electronics for you

Do-it-yourself

Sa Th ni eo

Image CompressIon Using Discrete Cosine Transform Technique Dr AhlAD KumAr

P

to take place. The software performs resented here is a MATLABthis process automatically as you based program for image comsend the file to one of your friends. pression using discrete cosine However, this process is visible only transform technique. It works for when the file to be transmitted is both coloured and grayscale images. large in size, such as the 5MB video Over the last few years, messaging in this example. apps like WhatsApp, Viber and Skype Image compression is the task of have become increasingly popular. representing an image with miniThese applications let s send and mum number of coefficients so that receive text messages and videos. the total memory occupied by the All of us make extensive use of these compressed image is much less than applications without knowing what the original image. With this reducactually goes behind the scene in tion of memory requirement for hightransmitting high-quality images and definition image, the transmission text. This article dwells on the image/ of these images onto the transmitvideo compression concept that is ting medium is much easier than being used by nearly all the Internetwithout compression. based messaging applications. In order to achieve the task of Fig. 1 shows screenshots of the image compression, it has to be repauthor’s mobile as it compresses the resented in a domain where high-defvideo using WhatsApp software. The inition images/videos are sparse. The application performs compression in two existing domains widely used in two steps: (a) preparing mode and (b) sending mode. The sending mode digital signal processing are spatial domain and frequency domain. basically deals with transmission of The third domain widely used data stream onto the communicanowadays in the field of image tion channel. So here we will restrict processing is sparse domain. In this ourselves to discussing the typical domain, mostly the coefficients are image/video compression algorithm that runs behind the preparing mode. sparse in nature, i.e., most of them The main job of the image/video compression algorithm is to reduce the size of the file to be transmitted. For example, in the case of a 5MB video file, the image/video compression software running behind the preparing mode in WhatsApp software makes the video smaller by up to 1MB, thus saving the memory Fig. 1: Compression process in mobile using WhatsApp software: (a) Application in preparing mode, (b) Application in sending mode space for transmission www.efymag.com

are zero with very few non-zero coefficients. Typically used techniques for transforming the spatial domain to sparse domain include wavelet, curvelet, singular value decomposition (SVD) and discrete cosine

(a)

(b)

(c) Fig. 2: (a) Original image, (b) Spatial domain (histogram) and (c) Sparse representation electronics for you | June 2017

129

Do-it-yourself

(a)

(b)

(c)

(d)

Fig. 3: (a) Original image and (b)-(d) compressed versions using DCT

Here, the compression ratios for the images shown in Figs 4(b)-(d) are given as 5, 6 and 7, respectively. It can be seen, as the compression ratio increases from left to right, the blocking artifacts tend to appear in an image. This can be clearly seen from Fig. 4(d).

Testing procedure

(a)

(b)

(c)

(d)

Fig. 4: (a) Original image and (b)-(d) compressed images

transform (DCT). The concept of sparse domain is illustrated in Fig. 2 in a simple way. Fig. 2(a) shows a high-definition original image that occupies 2.3MB of space. In spatial domain, this image is represented as a matrix of numbers, which are basically image-intensity levels. The plot of intensity levels, known as histogram of the image, is shown in Fig. 2(b). It can be observed from Fig. 2(b) that these intensity levels vary across a large range from 0 to 255. However, if you transform the same image using wavelet, curvelet, DCT or SVD domain, you get the plot of respective coefficients as shown in Fig. 2(c). It can be observed that the same image can be represented using fewer coefficients as most of the coefficients in these domains are nearly zero. Hence, discarding these nearly-zero coefficients and retaining only non-zero coefficients reduces the memory space required to store these coefficients, which, in turn, helps in compressing the image. Here, we use DCT for image compression. Please note, it is not known what algorithm WhatsApp software is utilising for compressing its images and videos. Our intention in this article is to present the underly130

June 2017 | electronics for you

ing concept behind the operation discussed in Fig. 1. Figure 3(a) shows a flower image that occupies 2.3MB of storage space. Using the DCT-based image compression algorithm, we obtained compressed images of sizes 392kB, 274kB and 223kB as shown in Figs 3(b)-(d), respectively. It can be seen, as the size of an image is compressed, artifacts tend to occur near the edges of the image. This is clearly visible in Fig. 3(d) where significant artifacts are visible. The compression ratio is defined as: K=Uncompressed size of an image/Compressed size of an image For the images in Figs. 3(b)-(d), the value of K is obtained as 5, 8 and 10, respectively. It can be observed from Fig. 3 that the images obtained after compression occupied less space and yet were good enough for visual inspection. Similar analysis is performed on grayscale image of a cameraman and the results are shown in Fig. 4.

efy Note

The source code of this project is included in this month’s EFY DVD and is also available for free at source. efymag.com

The program (code.m) can be used for colour and grayscale images both. Tulip.jpeg colour image and cameraman.jpeg grayscale images were used during the testing of this program. You need to select one image (either colour or grayscale) at a time. 1. Install MATLAB R2013a or later version in your system. Open the code.m file 2. If colour image is to be compressed, line number 11 of the code.m file has to be uncommented. It is already uncommented for the given program 3. If grayscale image is to be compressed, line number 14 of the code.m file has to be uncommented. 4. Once you have selected either step 2 or step 3, select Run command button. Then the program prompts you to enter the threshold value. After entering this value followed by pressing Enter key, you need to wait for some time till the compressed image pops up on the screen. 5. For a coloured image, choose any one threshold value (e.g., 5, 50 or 500) for generating images shown in Figs 3(b)-(d), respectively. 6. For grayscale images, choose one threshold value (e.g., 10, 60 or 100) for generating images shown in Figs 4(b)-(d), respectively.

Dr Ahlad Kumar is currently doing research in Concordia University, Montreal Canada in the field of image processing. His area of interest is in the field of image restoration. He received his PhD degree from University of Malaya, Malaysia in 2016. He is gold medalist for his M.Tech from ABV-IIITM in 2007

www.efymag.com

rs’ index Client name

Page No.

AN Wireless Solutions ...........................................................................145

Client name

Page No.

Green Vision Technologies .......................................................................3

Client name

Page No.

NMTronics India Pvt.Ltd. (www.nmtronics.com)......................................13

Analog Devices (www.analog.com) .........................................................27

GT Magnetics Pvt Ltd (www.gtmagnetic.com) ......................................136

No Smoke Oil ...........................................................................................55

Anant Enterprises (www.indiamart.com) ...............................................136

HARIHI Ohm Electronics .......................................................................134

OSRAM Opto Semiconductors (China) Co. Ltd. .......................................6

AqTronics Technologies Pvt Ltd ..............................................................19

HARISON Micro System........................................................................142

Pasternack .............................................................................................148

ASIAN Heating Alloys Industry ..............................................................139

Harwin Asia Pte Ltd..................................................................................31

Perfect Metal Works...............................................................................141

BIMM Technologies................................................................................142

Hicotronics Devices Pvt Ltd. .................................................................... 11

Perfect Radios........................................................................................141

Binay Opto Electronics Pvt Ltd (www.binayLED.com) ......................50, 51

Hind Electronika India Private Limited (www.hindelec.com) ..................45

Rajguru Electronics ................................................................................133

BlackT Electrotech .................................................................................141

HK Wentworth (India) Pvt Ltd ...............................................................147

REBUTOR Electronics Pvt Ltd ................................................................53

CADD Centre Training Services Private Limited ...................................135

Incubation Centre-Indian Institute of Technology Patna..........................21

RECOM Asia Pte Ltd ............................................................................. 117

C-DAC (Pune) ........................................................................................ 119

India Electronics Week 2017 ................................................................143

Rhydo Technologies P Ltd (www.rhydo.com)........................................131

Circuit Systems India Ltd ......................................................................103

infineon.....................................................................................................83

Rohde & Schwarz (www.rohde-schwarz.co.in) .......................................29

Digi-Key Electronics (www.digikey.com)....................................................2

Jai Mata Electronics ...............................................................................141

Sakthi Accumulators Private Ltd ............................................................142

Digital Promoters (I) Pvt Ltd ..................................................................141

Keysight Technologies India Pvt Ltd (www.keysight.com) ........................5

Schurter Electronics (I) Pvt Ltd ..............................................................137

Dinrack Integrated System Pvt Ltd ........................................................132

Kits‘n’Spares ............................................................................................91

Sealed Energy Systems ..........................................................................57

DSM India Private Limited ....................................................................134

LWI Electronics Inc. (www.livewireinfo.com) ...........................................23

Shavison Electronics Pvt Ltd (www.shavison.com)............................... 111

Dynalog (India) Ltd (www.dynadia.com) ...................................9, 121

MathWorks ...............................................................................................41

Shrey Plastic Moulders (www.shreyplasticmoulders.com)....................141

EFY Group: Subscription Form..........................................................70, 71

Matrix Comsec Pvt. Ltd............................................................................25

SIGLENT Technologies Co. Ltd...............................................................89

EFY University Program ....................................................................80, 81

Max Electronics N ........................................................................................

SINPRO Electronics Co. Ltd (www.sinpro.com)......................................63

Eita Technologies ...................................................................................142

Max Technology & Co. (www.maxtechnoloindia.com) ............................35

Softgrip Power Product LLP ..................................................................139

Electronic Assembly (www.lcd-module.de)..............................................50

Meco Meters Pvt Ltd (www.mecoinst.com) .................................65, 67, 69

ST Microelectronics Marketing Pvt Ltd ....................................................59

Element14 India Pvt Ltd.............................................................................1

Microchip Technology Hong Kong Ltd (www.microchip.com) .................33

Systellar Innovations (www.systellar.in).................................................138

Embedded Technology Labs (www.etechlabsindia.com)......................134

Minoo Impex...........................................................................................140

Testo India Pvt Ltd (www.testo.in)................................................... 115-117

FLIR Systems India Pvt. Ltd. (www.flir.com) ...........................................39

Montu Electronics LLP ...........................................................................141

Toshniwal Sensing Devices Private Limited ..........................................136

Fujitsu Electronics (Shanghai) Co. Ltd. ...................................................17

Mornsun Guangzhou Science & Technology Co. Ltd. ............................37

Transcom Instruments .............................................................................15

Fusion Power Systems (www.amptek.in) ..............................................132

Mo Electronics (India) Private Limited ................................................7

Universal Electronic Agencies (www.easternradio.co.in) ......................138

Future Hi-Tech Batteries Limited (www.fhtbl.com) ..................................57

MSS powertech pvt. ltd ..........................................................................142

VIGVEN Tech Mark Pvt Ltd (www.vigven.com).......................................61

Good Will Instrument Co. Ltd.................................................................146

National Controlling & Equipments ........................................................142

VIRSON Corporation .............................................................................141

Goodwin Electronics Pvt. Ltd. ................................................................141

NI Systems India Pvt Ltd .........................................................................43

Zhaoqing Beryl Electronic Co. Ltd. ..........................................................85

rs’ product categories index Products

Page No.

Automation & Robotics Dynalog (India) Ltd ....................................... 9, 121 HARISON Micro System .................................. 142 Batteries & Power Supplies Fusion Power Systems ..................................... 132 Future Hi-Tech Batteries Limited ....................... 57 Green Vision Technologies .................................. 3 Montu Electronics LLP ..................................... 141 Mornsun Guangzhou Science & Technology Co. Ltd. ......................................... 37 National Controlling & Equipments ................... 142

Products

Page No.

ST Microelectronics Marketing Pvt Ltd .............. 59 Zhaoqing Beryl Electronic Co. Ltd. ..................... 85 Consumer Electronics & Appliances BIMM Technologies ......................................... 142 BlackT Electrotech ........................................... 141 Display Systems Electronic Assembly .......................................... 50 EDA Tools (Including Deg & Drafting Aids)

Products

Page No.

PCBs, Assemblies & Sub Assemblies Circuit Systems India Ltd ................................. 103 Eita Technologies ............................................ 142 Goodwin Electronics Pvt Ltd............................. 141 Plugs, Sockets & Connectors Harwin Asia Pte Ltd ........................................... 31

Products

Page No.

FLIR Systems India Pvt Ltd. ............................... 39 Good Will Instrument Co. Ltd ........................... 146 Hind Electronika India Private Limited ................ 45 Keysight Technologies India Pvt Ltd. .................... 5 Meco Meters Pvt Ltd .............................. 65, 67, 69 Minoo Impex .................................................... 140 Rohde & Schwarz .............................................. 29

Pasternack ....................................................... 148

SIGLENT Technologies Co. Ltd .......................... 89

Reseller and Distributors

Transcom Instruments ....................................... 15

AqTronics Technologies Pvt Ltd ......................... 19 Digi-Key Electronics ............................................ 2

Testo India Pvt Ltd. .....................................115-117

Tools & Toolkits Rhydo Technologies P Ltd ................................ 131

MathWorks ........................................................ 41

Element14 India Pvt Ltd ....................................... 1

NI Systems India Pvt Ltd .................................... 43

HARIHI Ohm Electronics .................................. 134 Hicotronics Devices Pvt Ltd. ................................11

Training and Certification Institutes

Shavison Electronics Pvt Ltd. ............................111

Educational Training Kits

LWI Electronics Inc. ............................................ 23

CADD Centre Training Services

SINPRO Electronics Co. Ltd............................... 63

Kits‘n’Spares ...................................................... 91

Mo Electronics (India) Private Limited .......... 7

Private Limited .............................................. 135

Systellar Innovations ....................................... 138

Rhydo Technologies P Ltd ............................... 131

Perfect Radios ................................................. 141

C-DAC (Pune) ...................................................119

Rajguru Electronics ......................................... 133

Incubation Centre-Indian Institute of

Cabinets, Enclosures & Accessories

Industrial & Manufacturing Equipment

REBUTOR Electronics Pvt Ltd ........................... 53

DSM India Private Limited ............................... 134

Max Technology & Co. ....................................... 35

Universal Electronic Agencies ......................... 138

Jai Mata Electronics ........................................ 141

NMTronics India Pvt Ltd. .................................... 13

Shrey Plastic Moulders .................................... 141

VIRSON Corporation ........................................ 141

RECOM Asia Pte Ltd .........................................117 Sakthi Accumulators Private Ltd ...................... 142 Sealed Energy Systems .................................... 57

Transformers Sensors & Transducers

Anant Enterprises ............................................ 136

Toshniwal Sensing Devices Private Limited ..... 136

GT Magnetics Pvt Ltd. ...................................... 136 Wireless Modules/Modem

Components (Including Active & ive)

Materials (Including Chemicals &

Analog Devices................................................... 27

Consumables)

Services

Dinrack Integrated System Pvt Ltd .................. 132

DSM India Private Limited ............................... 134

Embedded Technology Labs ............................ 134

AN Wireless Solutions ..................................... 145 Embedded Technology Labs ........................... 134

Fujitsu Electronics (Shanghai) Co. Ltd. .............. 17 HK Wentworth (India) Pvt Ltd .......................... 147

Optics & Optoelectronics

Telecom Products

infineon .............................................................. 83

Binay Opto Electronics Pvt Ltd ..................... 50, 51

Matrix Comsec Pvt Ltd. ...................................... 25

Microchip Technology Hong Kong Ltd. ............... 33

OSRAM Opto Semiconductors (China)

MSS Powertech Pvt Ltd.................................... 142

Technology Patna ........................................... 21

Co. Ltd ............................................................... 6

Wires & Cables ASIAN Heating Alloys Industry ......................... 139

Test & Measurement Equipment (Including

Perfect Metal Works ......................................... 141

Softgrip Power Product LLP ............................. 139

Indicators & Monitors)

Miscellaneous

Schurter Electronics (I) Pvt Ltd ......................... 137

Systellar Innovations ....................................... 138

Digital Promoters (I) Pvt Ltd ............................. 141

No Smoke Oil ..................................................... 55

144

June 2017 | electronics For You

www.eFYmag.com

EFY Magazine Attractions During 2017 Month

tEchnologY Focus

ElEctronics DEsign

MArkEt survEYs

January

Computer Vision

Lowering Power Consumption

Electronic Component Manufacturing

February

Smart Fabrics

Improving Wireless Signal Performance

Electronics Manufacturing Services

March

Exciting Technologies Powering the IoT

Building More Reliable Printed Circuit Boards

Industry Outlook for 2017-18

April

Virtual and Augmented Reality

Developer Boards: DIY and Hobbyist Applications

Printed Circuit Boards

May

Smart Robotics

Fight of the Processors: Ultra-Low-Voltage Computing The Internet of Things (Mobile and Portable Devices)

June

Artificial Intelligence

How to Get the Best Design for Manufacturing

Strategic Electronics

July

5G and Beyond

Developer Boards: Industrial Applications

Mobile Handset and Telecom Device Manufacturing

August

3D Printing

Improving Thermal Dissipation

LED Lighting

September Industrial IoT

Which Input Technologies Should You Use

Solar

October

The Brains of Mobile Devices

Ruggedising Hardware

Electronics Manufacturing Equipment (Both SMT and Non-SMT Categories)

November

Security of the IoT

Fight of the Processors: High-Performance Computing Test and Measurement

December

Implantable and Edible Electronics

Which Output Technologies Should You Use

www.efymag.com

Materials and Chemicals

electronics for you | June 2017

145