Distributed Resampling Gaussian Particle Filtering for Heterogeneous Networks

In this work, we propose a fully distributed resampling Gaussian particle filtering (D-ReGPF) algorithm for heterogeneous networks, where each sensor could implement distinct adaptation rules. In order to enhance the robustness of the algorithm, we further develop an adaptive combination strategy. Illustrative simulations of the direct target tracking problem involving heterogeneous networks validate that the proposed D-ReGPF incorporated with the adaptive combiners could markedly outperform the existing distributed Gaussian particle filtering with conventional combiners for homogeneous networks.