A sensor selection approach for target tracking in sensor networks with quantized measurements

Abstract

This paper extends our earlier work on sensor selection [1]. We are now focusing on a more challenging problem of how to effectively utilize quantized sensor data for target tracking in sensor networks by considering sensor selection problems with quantized data. A subset of sensors are dynamically selected to optimize the tracking performance. The one-step- look-ahead posterior Cramer-Rao Lower Bound (CRLB) on the state estimation error is proposed as the sensor selection criterion. Particle filtering method is employed to compute the posterior CRLB, as well as to estimate the target state. Simulation results show that the proposed posterior CRLB based method outperforms the one based on information theoretic measures.