Using Spark Fund Award to Reduce Power Consumption of IoT Devices

Aatmesh Shrivastava

ECE Assistant Professor Aatmesh Shrivastava is using the Spark Funds he received earlier this year to extend the battery life of devices connected to the Internet of Things.

Using innovative circuit design to extend the battery life of the IoT — with Spark Fund awardee Dr. Aatmesh Shrivastava

Do you ever try to use your smartwatch to record a long workout or hike, only to have it give a low battery warning halfway through? Or have you ever run down your smartphone battery trying to connect to all of your smart home devices like your smart thermostat or home security system?

DSC 5223
Photo by Joshua Brown

You’re not alone! With the advent of smart devices, things like our homes, cars, and offices are all more connected than ever via the Internet of Things (IoT) — but with this connectivity comes a big drain on our batteries. Most IoT devices are not always connected and save power by only connecting when the device is in use or at required intervals, but they still consume a lot of power.

Dr. Aatmesh Shrivastava and his team of graduate student researchers have spent the last three years at Northeastern developing circuits that can reduce the power consumption of radios and wireless communication within IoT devices by a few orders of magnitude. This reduction in power consumption would enable IoT devices to be continuously connected and enhance usability. With an estimated 75-billion devices expected to be connected through IoT networks by 2025, this technological advancement could open new possibilities for the IoT.

Reducing the power consumption of IoT devices with improved circuit design

Power consumption is one of the biggest technological challenges for IoT technology. Instead of being continuously connected, most IoT devices use Wake Up Radio (WUR) technology to save power. WUR technology wakes up the primary radio in the IoT device only when it is in use to cut down on power consumption.

However, even with the power-saving WUR technology, things like channel noise, interference, and random fluctuations, among other issues, can cause excess power consumption for IoT devices. With the growth rate of the IoT, this amount of power waste is simply not acceptable.

Aatmesh TechDr. Shrivastava and his team have come up with a new chip-architecture for WURs that is resilient to the channel noise and interference. This new design achieves a million-times reduction in power consumption over conventional radios. The technology is developed using high-precision, ultra-low power analog circuit design technology that was developed by Dr. Shrivastava.

Using his technology, the team aims to develop a wake-up radio circuit that uses less than 20 nano-watts, can achieve a sensitivity greater than -90 dBm, and has a connectivity range of approximately 100-feet. The radio architecture is based on the energy detection of the incoming radio signal implemented using high sensitivity passive energy detection circuits to realize ultra-low-power operation.

If this technology is successfully commercialized, the time in between needing to charge your cell phone batteries could change from daily to every couple of weeks.

Commercialization through The Spark Fund

Dr. Shrivastava’s lab was awarded a 2022 Spark Fund grant, which will be used to advance the commercial prospects of their research.

“We are using the CRI’s Spark Fund to develop the next generation of radio technology and to produce a new demo to raise our next level of funding,” said Dr. Shrivastava.

The CRI helped Dr. Shrivastava and his team with patent filing, commercialization and incorporation, and an SBIR write-up. Several of his inventions, which are at different stages of prosecution and grant with USPTO, have been filed and protected through patent applications by the CRI.

“The CRI promises to help us meet our goals to realize a truly disruptive technology,” said Dr. Shrivastava.

Source: Center for Research Innovation

Related Faculty: Aatmesh Shrivastava

Related Departments:Electrical & Computer Engineering