Adaptive Sensing and Estimation of Sparse Signals
Oct 25, 2012
from 11:00 AM to 12:30 PM
|Where||Engr. IV Bldg., Maxwell Room 57-124|
|Contact Name||Prof. van der Schaar|
|Add event to calendar||
University of Michigan
Adaptive sensing and inference have been gaining interest in recent years in signal processing and related fields. In this talk, I discuss the sequential adaptive estimation of sparse signals under a constraint on total sensing resources. The advantage of adaptivity in this context is the ability to focus more resources on regions of space where signal components exist, thereby improving the signal-to-noise ratio. A dynamic programming formulation is derived for the allocation of sensing effort to minimize the expected estimation loss. Based on the method of open-loop feedback control, allocation policies are then developed for a variety of loss functions. The policies are optimal in the two-stage case and improve monotonically thereafter with the number of stages. Numerical simulations show gains up to several dB as compared to recently proposed adaptive methods, and dramatic gains approaching the oracle limit compared to non-adaptive estimation. An application to radar imaging is also presented.
Dennis Wei received S.B. degrees in electrical engineering and in physics in 2006, the M.Eng. degree in electrical engineering in 2007, and the Ph.D. degree in electrical engineering in 2011, all from MIT. He is currently a post-doctoral researcher in the Department of Electrical Engineering and Computer Science at the University of Michigan. His research interests lie broadly in signal processing, optimization, and statistical inference and learning, with a current focus on adaptive sensing and processing. He has also worked on the design of sparse discrete-time filters. Dr. Wei was a recipient of the William Asbjornsen Albert Memorial Fellowship at MIT and a Siebel Scholar.