Enabling Low-power Computation, Sensing, and Communication for the Internet-of-Things
Internet-of-Things (IoT) devices serve as gateways connecting the digital world to the physical world. Although we already have powerful tools to understand various data in the digital world, IoT devices are currently insufficient for capturing and processing large amounts of physical world data. In this talk, I will present two systems we built to address this problem. I will first present a low-power backscatter radio system designed for IoT. Our radio consumes 100~10x less power compared to existing WiFi radios. It does so by reflecting existing WiFi traffic in the air while embedding data. Then, I will present a runtime system that supports running large software tasks on intermittently powered computing platforms even when such platforms have tiny energy buffers. We demonstrate that our system significantly improves the operational regime of intermittently powered computing platforms.
Pengyu Zhang is a postdoc researcher at Stanford University. He received his Ph.D. from UMass Amherst and his bachelor’s from Tsinghua University. He works on designing and building low-power embedded systems and wireless systems to efficiently support applications for the Internet-of-Things. He received the ACM SIGMOBILE Doctoral Dissertation award, and best paper award nominations at MobiSys 2017, SenSys 2016, UbiComp 2016 and MobiCom 2014.