Exploiting Circuit Nonlinearity for Biological Signal Processing
Kofi Odame, Electrical and Computer Engineering Department, Georgia Institute of Technology
Friday, February 1, 2008
Quicktime Large (48 MB)
Quicktime Small (34 MB)
MP3 (22 MB)
This seminar is part of the Jones Seminars on Science, Technology, and Society series
Nonlinear processing is often more suitable than the traditional linear approach is for analyzing biological signals. Unfortunately, digital nonlinear operations are computationally expensive. In contrast, a large variety of nonlinear operations can efficiently be implemented in analog electronics, operating at real-time speeds. The low level of accuracy generally associated with analog processing is not a concern in this scenario, as biological signals themselves typically have low signal-to-noise ratios. In this talk, I suggest that extremely low-power nonlinear processing is possible by shunning the binary perspective of digital, as well as the linearized view common to analog, in favor of a more genuine representation of the transistor that allows rich, nonlinear behavior to be fully exploited. The highly-efficient processors that result are necessary for battery-powered applications like fully-implanted cochlear prosthetics.
About the Speaker
Kofi Odame is a PhD Candidate at Georgia Institute of Technology's Electrical and Computer Engineering Department. His fields of concentration are analog VLSI design and nonlinear dynamical systems theory. His research interests are in low-distortion amplifiers and filters, as well as in novel circuits for nonlinear signal processing. Prior to Georgia Tech, he was at Cornell University, where he earned the B.Sc. ('02) and M.Sc. ('04) degrees in Electrical and Computer Engineering, with minors in Computer Science.