Consensus-based Distributed Target Tracking in the Presence of Active False Targets

Multi-station radar fusion of detecting and tracking targets is currently a promising research direction, especially in the presence of active false targets. However, the existing researches are mainly based on the centralized radar network. In this regard, this paper proposes a target tracking algorithm based on data-level fusion in distributed network. First, unscented Kalman filter (UKF) algorithm is used to generate local state estimations. Second, according to the spatial position correlation of real targets, all of the local estimated values from multiple stations are associated, which is aimed to eliminate false targets by nearest neighbor (NN) algorithm. Third, covariance intersection (CI) algorithm and consensus algorithm are combined, named as consensus CI algorithm, in order to track the real target and improve the fusion accuracy. Finally, the effectiveness of the algorithm is verified through simulation.