PSO-PF target tracking in range-based Wireless Sensor Networks with distance-dependent measurement noise

In this paper a Particle Swarm Optimization (PSO) based Particle Filter (PF) for tracking a rotating object in a range-based Wireless Sensor Network (WSN) equipped with distance measuring sensors is developed. The distance-dependent measurement error is incorporated in the observation equation as a multiplicative noise. To overcome the impoverishment problem of PF, weighted aggregation of the likelihood and the prior is maximized through PSO in order to move the prior samples towards regions of the state space where both the likelihood and the prior are significant. Performance of the proposed approach is compared with that of Extended Kalman Filter (EKF) state estimator. Simulation results show the effectiveness of the developed target tracking approach.