You are here: Home / Content / 2016 / Anantha Chandrakasan: “We will even be able to power a chip with energy from our body”

Anantha Chandrakasan: “We will even be able to power a chip with energy from our body”

27 January 2016 - American engineer and computer scientist Anantha P. Chandrakasan (47) is a world authority on electronic chip design. He’s been doing research into energy-efficient circuits and systems for twenty-five years. With the Internet of Things, whereby practically all devices exchange information without human intervention, we’ll need more and more technologies that function without electricity or batteries.

27 January 2016 - American engineer and computer scientist Anantha P. Chandrakasan (47) is a world authority on electronic chip design. He’s been doing research into energy-efficient circuits and systems for twenty-five years. With the Internet of Things, whereby practically all devices exchange information without human intervention, we’ll need more and more technologies that function without electricity or batteries.
Submitted by Anantha P. Chandrakasan

“My mother was a biochemist. Seeing her work in her lab has definitely inspired me to pursue an academic career. Why I didn’t go for biochemistry? Well, I was already fascinated by electronics at the time: building chips and electronic gadgets was the exciting thing back then, just like writing software and apps is today”, says Professor Anantha Chandrakasan.

His decision proved to be successful. By the end of 2015, he was not only elected member of the National Academy of Engineering, an association of American top engineers, but also learned that he would be awarded his first honorary doctorate in Leuven. “I am very honoured. I already know Leuven a little bit. I visited the city a couple of times for a PhD defence or for a visit to Imec, the Leuven-based research institute for nanoelectronics.”

For every engineer and engineering student with an interest in microelectronics, the name ‘Chandrakasan’ should ring a bell. His publications are used as textbooks, his papers are must-reads. Chandrakasan heads the Electrical Engineering and Computer Science Department at the prestigious Massachusetts Institute of Technology (MIT). He may be even more famous as the driving force behind the annual International Solid-State Circuit Conference (ISSCC), the place to be for industrial players and academics from the field of microelectronics.

Moore’s law

In recent years, the end of Moore’s law has been predicted numerous times already, because the nanoscale level comes with certain physical limits. For one thing, transistors can no longer be switched off, similar to a tap that can no longer be completely turned off once it’s been turned on. This causes the energy loss in standby modus. The limits to Moore’s law are a challenge to designers: how do you design a chip with a few billion transistors that performs as quickly as possible and uses as little energy as possible? Chandrakasan’s specialty is exactly that last aspect: “I did my PhD at the University of California, Berkeley, under supervision of Professor Robert Brodersen, who is a pioneer in the field of energy-efficient circuits and systems.” Chandrakasan refers to a – by now iconic – paper that he published in 1992 together with Brodersen and colleague Samuel Sheng: “It was about the factors that have an impact on energy consumption and about how you can save energy. When we published the paper, hardly anyone was paying attention to this issue. But because the article was so well received, energy was soon on the agenda.”To understand Chandrakasan’s work, we have to return to 1965. That is when Gordon Moore, one of the founders of chip producer Intel, predicted that the number of transistors per integrated circuit – or, in plain English, per chip – would double every eighteen months because of their increasingly small size. The prediction became known as ‘Moore’s law’. Even though the eighteen-month time span proved somewhat off, Moore’s law became reality. In the past fifty years, chips have become increasingly small and intelligent. Computers now have chips with two billion transistors – switches, so you like – on just a few square centimetres. This progress has resulted in a boom of consumer electronics: laptops, mobile phones, tablets, charging devices, ...

The low energy use of chips is even more important today, especially in the context of the Internet of Things. More and more objects around us are online: our smartphone, the watch that also monitors our heart rate, the thermostat and the fridge in our home, our car, traffic lights, ... Everything contains chips that can measure something and share that information with all these other devices – wirelessly and without human intervention. But all these chips obviously use energy as well, and connecting millions of chips to the electricity grid is not an option. Batteries are a possible alternative, but, unfortunately, battery technology has not kept pace with the progress of chips.

Harvesting energy

Chandrakasan’s research group at MIT studies, among other things, energy harvesting: ways to make sensors and chips draw energy from their environment, so that they can recharge their tiny battery or even function without any battery at all. “The best-known examples are solar panels and self-winding watches.” Just a couple of years ago, Chandrakasan’s team developed a chip that simultaneously gets its energy from sunlight, vibrations, and heat. “But we’re also looking into less conventional sources of energy, such as our own body. After all, the human body generates electricity as well. Our so-called biological battery is located deep within our inner ear. We have shown that it is possible to tap energy from that battery for electronic circuits. That is my vision. Ultimately, we should be able to make a chip that is so energy efficient that we can power it from ambient energy sources or even energy from our own body.”

Chandrakasan is not a technology buff who only pays attention to the technical aspects of his work. “Moore’s law proved right: we can make very cheap and energy-efficient transistors. Now our job is much more about the exciting applications we can build with them.” He talks about his research into biomedical applications with great enthusiasm: “A domain in which we’re really moving forward is the production of wearable and implantable biomedical electronics. In collaboration with the Andrea Bocelli Foundation – founded by the blind, Italian tenor and his family – we’ve recently developed a microchip that the visually impaired can wear as a pendant. The camera on the pendant measures the distance between the person and any obstacles. The chip translates that information via a braille device, so that the wearer knows how far removed the obstacle is. Another example is a fully implantable cochlear implant for the hearing impaired or the deaf: there is a sensor in the cochlea of the ear that can wirelessly recharge itself, so that you no longer need an external part like you do with a traditional cochlear implant.”

So when can we expect those applications? “I get that same question via email almost every day: when will these devices be ready for use? For the wearable applications, the process may be quick. The implants will take a couple more years of doing further research and getting approvals. But this is a point I want to make: academic researchers can speed up that process themselves. And educational institutions have to integrate the knowledge on that issue in their curriculum. I’m teaching a workshop for students on start-ups, innovation, and entrepreneurship. As academics, we cannot take it for granted that research innovations automatically get translated into commercial products.”

Who is Anantha P. Chandrakasan? 

  • ° 1968 in India
  • Studied electrical engineering and computer sciences
  • Completed his PhD at the University of California, Berkeley, in 1994
  • Married; three children
  • Professor and head of the Electrical Engineering and Computer Science Department at the Massachusetts Institute of Technology (MIT)
  • Conference chair of the annual International Solid-State Circuit Conference (ISSCC)

Ilse Frederickx & Katrien Bollen