Distributed tracking in AD-HOC sensor networks

In this paper, we present distributed algorithms for tracking a moving source via an ad-hoc network of sensors. Tracking is performed by employing a Kalman filter at all detecting nodes in the network. The Kalman filter employed at any given node exploits the availability of source-location snapshot and prediction estimates, both of which are computed via distributed locally constructed algorithms over the ad-hoc network. As our brief simulation-based analysis reveals, the source-tracking performance of the proposed algorithms is a function of the motion dynamics of the source, the snapshot source-localization algorithm employed, the network topology, and the number of iterations employed in the distributed approximation algorithm