Distributed multi-sensor tracking in wireless networks using nonparametric variant of sum-product algorithm

Graphical models have been widely applied in solving distributed inference problems in wireless sensor networks (WSNs). In this paper, we formulate the distributed multi-sensor tracking problem in a WSN as an inference problem on a factor graph. Using particle filtering methods, we propose a nonparametric variant of sum-product algorithm (SPA), called sequential particle-based SPA (SPSPA), for factor graphs to infer the multi-sensor target states over time. In the proposed algorithm, importance sampling methods are used to sample from message products, and the computational complexity of SPSPA is thus linear in the number of particles. We apply the SPSPA to a distributed multi-sensor tracking problem, and evaluate its performance in terms of the measurement noise and the number of particles.