Back to the Future: Analog Signal Processing

Prof. Boris Murmann
Stanford University

Abstract:
In the early days of my engineering career, analog circuit design was viewed as "doomed" and no longer needed in future systems that will be almost entirely digital and rely on "standard" analog interfaces that need no further optimization. A we all know, this was wrong. Nonetheless, it is true that in most modern systems, the analog interface often plays only a minor role in the overall signal processing function. This is evidenced, for example, by our constant craving for "software defined" devices in which "all" available information is dumped into the digital domain and sorted out using purely digital processing.

In this talk, I will review a number of research ideas in which analog blocks participate more actively in the acquisition, selection and processing of the desired information. We will discuss concepts related to "fooling Nyquist" and extracting desired analog-domain information using low-rate and low-bandwidth observations. In addition, we will investigate the potential for mixed-signal co-processors in machine learning algorithms and other applications that are amenable to "approximate computing."

Bio:
Boris Murmann joined Stanford University in 2004, where he presently serves as an Associate Professor of Electrical Engineering. He received the Ph.D. degree in electrical engineering from the University of California at Berkeley in 2003. From 1994 to 1997, he was with Neutron Microelectronics, Germany, where he developed low-power and smart-power ASICs in automotive CMOS technology. Dr. Murmann’s research interests are in the area of mixed-signal integrated circuit design, with special emphasis on data converters and sensor interfaces.