INDUSTRIAL ENGINEERING AND OPERATIONS RESEARCH

PRESENTS

IEOR MONDAY SEMINAR

MAY 8, 2006

Sensor Redundancy and Robust Estimation

Yu Ding

Assistant Professor

Dept. of Industrial and Systems Engineering

Texas A&M University

Abstract:

Smart devices with multiple on-board sensors, networked through wired or wireless links and deployable in large numbers, are distributed in physical systems and environments for broad applications ranging from security surveillance, healthcare delivery, transportation control, to manufacturing quality control. In order to allow the potential of distributed sensor systems to be fully realized, one crucial aspect is to provide robust methods to ensure reliable operations of sensor systems in the presence of potential sensor failures. Existence of sensor redundancy is one of the fundamental reasons enabling the robustness or fault tolerance capability for a sensor system. This talk presents a redundancy analysis that quantifies the effect of redundancy on system robustness. A bound-and-decompose algorithm is devised to determine the degree of sensor redundancy. The talk will show that the sensor redundancy measure is related to the breakdown point measure used in robust regression. The applications of the resulting sensor redundancy measure may be in the following aspects: it can quantify the fault tolerance capability of a sensor system, help select proper parameters to be used in a robust estimator, or compare different sensor system designs and guide their future design toward a robust system.

Bio:

Dr. Yu Ding received a B.S. degree from the University of Science and Technology of China in 1993, a M.S. degree from Tsinghua University in 1996, the second M.S. degree from Penn State University in 1998, and Ph.D. degree from the University of Michigan in 2001. Dr. Ding is currently an assistant professor in the Department of Industrial and Systems Engineering at Texas A&M University. His research interests are in the general area of applied statistics and quality engineering, with emphases on data-mining methods for analysis and design, and optimal utilization of distributed sensor systems. His current research is sponsored by National Science Foundation, Nokia, Motorola, and the State of Texas Higher Education Coordinating Board. Dr. Ding received NSF CAREER Award in 2004 and an IIE Transactions Best Paper Award in 2006. Dr. Ding currently serves as a department editor for IIE Transactions on Quality and Reliability Engineering and as an associate editor for IEEE Transactions on Automation Science and Engineering.

TIME AND LOCATION: 3:30 - 4:30 P.M. - 3108 ETCHEVERRY HALL

REFRESHMENTS WILL BE SERVED @ 3:00 P.M.