CAREER: Digital Signal Conditioning Techniques to Improve Integrated Circuit Design Performance

University Of California-Los Angeles, Los Angeles CA

Investigators

Abstract

The objective of this research is to develop high performance analog/mixed-signal integrated circuits that are crucial to the advancement of wireless communication, sensor network, health, and biotech applications. The approach is to use signal processing algorithms implemented in digital logic to subtly modify the statistical properties of the analog signals such that inevitable transistor nonlinearity and manufacturing process variability do not degrade circuit performance. The intellectual merit is that it is a promising, fundamentally different approach that does not depend on brute circuit optimization but instead applies elegant signal processing principles to improve circuit performance. It trades off algorithmic complexity for relaxed analog circuit design thereby reducing power consumption, die area, design time, and offering higher performance and fabrication yields. The broader impacts of the project are three fold. It will help perpetuate California's position as a technology leader in the wireless communication, sensor network, and consumer electronics industries. Second, it will encourage students from underrepresented groups to take up advanced engineering careers by providing graduate students and summer interns recruited specifically from such groups to work on the proposed project. Third, the educational activity addresses the need for a change in the philosophy of circuit design education. By training future circuit designers in the fundamental interplay between circuit design practice and signal processing theory, it will better prepare them to face the immediate challenges of integrated circuit design, and to adapt to its impending evolution from microelectronics into newer technologies.