You are here

C3 Guest Lecture Series

Data-Efficient Quickest Change Detection
Venu Veeravalli
Prof. Veeravalli received the Ph.D. degree in Electrical Engineering from the University of Illinois at Urbana-Champaign in 1992, the M.S. degree from Carnegie-Mellon University in 1987, and the B.Tech degree from Indian Institute of Technology, Bombay in 1985. He is currently a Professor in the department of Electrical and Computer Engineering (ECE) and the Coordinated Science Laboratory (CSL) at the University of Illinois at Urbana-Champaign. He was on the faculty of the School of ECE at Cornell University before he joined Illinois in 2000. He served as a program director for communications research at the U.S. National Science Foundation in Arlington, VA during 2003-2005. His research interests include wireless communications, distributed sensor systems and networks, detection and estimation theory, and information theory. He is a Fellow of the IEEE, and a recipient of the IEEE Browder J. Thompson Best Paper Award and the U.S. Presidential Early Career Award for Scientists and Engineers (PECASE). He served as a distinguished lecturer for the IEEE Signal Processing Society during 2010-2011.
In the classical quickest change detection problem, there is a sequence of observations whose distribution changes at an unknown time, and the goal is to detect the change as quickly as possible, subject to a false alarm constraint. In many engineering applications of quickest change detection, there may be a cost (e.g., energy) associated with acquiring observations. We therefore consider the quickest change detection problem with an additional constraint on the cost of the observations used in the detection process, i.e., we seek algorithms that are data-efficient or energy-efficient. The objective is to select an observation control policy along with the stopping time at which the change is declared, so as to minimize the average detection delay, subject to both a false alarm constraint as well as a constraint on the average number of observations used before the change point. We consider both Bayesian and minimax formulations of the problem, and extensions to the setting where the observations are available at a set of distributed sensors. In all of these cases, we develop algorithms that are first-order asymptotically optimal as the false alarm rate goes to zero. We also demonstrate through numerical studies that our algorithms can be considerably more efficient than algorithms based on fractional sampling, where the observations to be skipped are determined a priori in order to meet the observation constraint. (This is joint work with Taposh Banerjee.)
January 11, 2013
234 Larsen

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer