Distributed Estimation in Sensor Networks
by
Jinjun Xiao
Washington University in St. Louis
Coauthors: Zhi-Quan Luo, UMN; Arye Nehorai, WUSTL

We will provide an overview of distributed compression-estimation in wireless sensor networks. A general formulation of distributed estimation under rate constraints will be introduced, and pertinent signal processing algorithms will be discussed. We will focus on rate-constrained distributed estimation for various signal models with variable knowledge of the underlying data distributions. We show that for a large number of cases, distributed estimation is possible with small bandwidth requirement and negligible performance degradation when compared to the corresponding centralized estimators. We will also present that for estimation in inhomogeneous environments, resource allocation and sensor scheduling algorithms can result in considerable performance improvement.

Date received: March 21, 2007