Group sparsity based multi-target tracking in multi-static passive radar systems using Doppler-only measurements

In this paper, we consider the problem of multi-target tracking in a multi-static passive radar system using Doppler-only measurements. In a multi-static configuration, the observability and estimation accuracy of target states can be significantly improved by simultaneously exploiting all available measurements. Track-before-fuse and fuse-before-track are the two fusion paradigms proposed in the literature to utilize such multi-static measurements. The fuse-before-track approach involves a minimal information loss and thus achieves a better accuracy and robustness than the track-before-fuse counterpart. However, despite the obvious advantages in terms of estimation accuracy and robustness, the centralized measurement fusion approach is difficult due to the prohibitive computational cost. As such, the track-before-fuse approach has been commonly used in multi-static passive radar tracking systems using Doppler-only measurements. In this paper, we exploit a group-sparsity based algorithm to simultaneously utilize the Doppler shift measurements at all bistatic pairs to obtain the target state estimates directly in Cartesian coordinate system. The estimated target states at each sampling instant are then fed as the inputs to the linear Gaussian mixture probability hypothesis filter, which removes the false measurements and correctly associates the measurements to the respective targets. Simulation results are provided to validate the ability of the proposed method to successfully handle the multi-target tracking problem in a challenging environment characterized by missed detection and false measurements.