Event-Based Recursive Distributed Filtering

Abstract

An important practical problem with the wireless sensor networks is how to find distributed estimators or filters to extract the information about the state vectors of the target plants from observations contaminated with external disturbances. It is generally known that the traditional Kalman filter algorithm is a recursive least mean square (LMS) one dealing with a single node and is optimal for linear systems with exact system models. On the other hand, to make use of the spatial information of the sensor nodes, distributed filtering problems have recently gained much research attention.