2018年11月19日上午9：30，美国斯坦福大学Boris Murmann教授应邀访问微纳电子学研究院，在微纳电子大厦103报告厅做了题为 “Mixed-Signal Circuit Design for the Data-Driven World”的学术报告。
Mixed-Signal Circuit Design for the Data-Driven World
Mixed-signal interfaces are essential bridges between the physical world and the digital information processing backbone. After many decades of incremental circuit-level refinements, innovation in such interfaces is increasingly driven by application-level insight and system-specific customizations. As a result, the traditional building block boundaries are blurring, and the extraction of information occurs through symbiotic interplay between analog and digital signal processing. In this talk, I will illustrate these trends through examples from my group’s research portfolio. In a first set of projects, we are looking at interfaces that “fool Nyquist” and lead to information-centric and data compressive digitization. A second area of interest is embedded machine learning. With the growing need for always-on operation, we are considering solutions that are tightly coupled to the sensor front-end and exploit mixed-signal techniques to reduce the data volume close to its source. In addition, we explore mixed-signal techniques to lower inference energy within mostly-digital neural network accelerators.