A collaborative target tracking framework using particle filter

Wireless sensor networks and communication technologies have dramatically revolutionized our lives and have hitherto spurred a host of applications in fields like surveillance and target tracking. Collaborative sensor networks largely depend on statistical estimation theory to estimate the target trajectory. Recent research has reported the use of a variety of filters - both linear and nonlinear in estimation and tracking problems. In this paper we propose a collaborative target tracking framework which uses a particle filter and a Mutual Information Sensor Selection Algorithm for target tracking. The major objective is to prolong the network lifetime and reduce the energy expended by the sensors using the most informative subset of sensors to provide non redundant data in the fusion process. The performance is evaluated using metrics like tracking energy, tracking quality, mean squared error and root mean squared error.