An improved particle filter algorithm for target tracking in binary wireless sensor networks

Abstract

An improved particle filter algorithm with adaptive process noise variance is proposed for target tracking in binary wireless sensor networks (BWSN). The standard particle filter (SPF) is not fit for target's constant velocity model and has cumulative effect of prediction error. The proposed algorithm adopts real-time updating process noise variance to eliminate the cumulative effect of particle filter prediction error. The simulation results show that the proposed algorithm is suited for target's constant and variable velocity model with higher accuracy than SPF.