Distributed Tracking with consensus on noisy time-varying graphs with incomplete data

In this paper we address the problem of distributed tracking with consensus on a time-varying graph with incomplete data and noisy communication links. We develop a distributed and collaborative tracking with consensus algorithm by combining distributed Kaiman filtering with consensus updates to handle a time-varying network topology in which not every node has local observations to generate own local tracking estimates. We introduce the concepts of active node set and connectivity graph to characterize such a network, and by merging these two, an effective network graph is obtained. Simulation results and performance analysis of the proposed algorithm are given and compared with that of distributed local Kaiman filtering with centralized fusion.