Simulation of Localized Performance-Guided Reconfiguration of Distributed Sensor Networks

Abstract

This paper describes a localized Performance-Guided Reconfiguration (PGR) algorithm for reconfiguring a distributed sensor network in a target localization scenario using binary sensor data. PGR works by activating redundant sensors available to replace failed sensors based on the improvement in target localization performance offered. Localized application of PGR helps to achieve globally optimal network performance. The target localization performance of a distributed sensor network was compared for PGR, reconfiguration by activating the redundant sensor nearest the failed sensor, and no reconfiguration. Localized PGR was superior to both other approaches, and always guaranteed the desired level of performance.