Date: April 3, 2020
Speaker: Xiyuan Tang
Faculty host: Prof. Mingoo Seok
Abstract: The artificial intelligence (AI) revolution has led to data explosion. As predicted, global data traffic will grow at approximately 30% each year. Given thermal and energy considerations, this exponentially increased data traffic implies a huge demand for highly energy-efficient high-speed data links. Since this massive amount of data has to be harvested by billions of sensor front ends, the battery life of the sensors becomes the most critical challenge to IoT growth, in turn imposing stringent energy constraints for signal conditioning circuits. In this talk, I will first present integrated circuit techniques that advance the energy-efficiency of the state-of-the-art data acquisition systems. Subsequently, I will introduce a low-cost reference stabilization technique that significantly relaxes the reference settling requirement for high-speed SAR ADCs. Finally, in addition to circuit techniques, I will also unveil latest developments in AI-assisted analog layout. By leveraging computer vision techniques, the trained AI can mimic an engineer’s behavior and produce high-quality analog layouts.
Bio: Xiyuan Tang received the B.Sc. degree (Hons.) from the School of Microelectronics, Shanghai Jiao Tong University, Shanghai, China, in 2012, and the M.S. and Ph.D. degree in electrical engineering from The University of Texas at Austin, Austin, TX, USA, in 2014 and 2019 respectively, where he is currently a post-doctoral researcher. He was a Design Engineer with Silicon Laboratories, Austin, from 2015 to 2017, where he was involved in the receiver design. The intellectual focus of his future research is on developing integrated circuit solutions to advance the development in emerging technologies, with focus on IoT (e.g., intelligent sensor), healthcare (e.g., low-cost monitor/diagnosis), and next-generation communication (e.g., 5G). Dr. Tang was a recipient of the IEEE Solid-State Circuits Society Rising Stars in 2020 and Silicon Labs Tech Symposium Best Paper Award in 2016.