Abstract: This article proposes an unscented-mixed centralized particle filter (UM-CPF) to deal with multi-sensor fusion based target tracking applications in wireless sensor networks (WSNs). The algorithm utilizes a mixed particle-propagation scheme. In the process of multi-sensor measurement fusion using a centralized particle filter (CPF) in WSN, a certain number of particles in the CPF are propagated by using a Gaussian distribution obtained from an unscented transformation (UT) as the proposal distribution, while the rest of the particles are simply propagated by using the state transition prior distribution. Simulation results of fusion-tracking in WSN show that, with a similar simulation speed, the mixed filter significantly increases the tracking accuracy and robustness as compared with the results obtained by the pure CPF.

Key words: wireless sensor network, target tracking, particle filter, multisensor fusion, mixed filtering