Tracking Multiple Targets in MIMO Radar Via Adaptive Asymmetric Joint Diagonalization

Direction finding for multiple targets is one of the most important tasks in MIMO radar applications. The existing algorithms such as MUSIC and ESPRIT for direction finding can only achieve the localization of multiple targets with unchanged location parameters due to their batch mode. In this paper, we show that the multiple targets localization and tracking in a bistatic MIMO radar system can be solved by adaptive asymmetric joint diagonalization, where the fluctuation of target radar cross section (RCS) is regarded as a source of time diversity for matched data, and the algebraic structure of the matched data themselves rather than their statistics is exploited. The resulting algorithm can achieve not only the localization but also tracking of multiple targets since it estimates the target location parameters in a recursive mode. The identifiability of the proposed algorithm is analyzed and the Cramer–Rao lower bound is also derived as performance benchmark. Simulation results show that the proposed algorithm has comparable localization accuracy to batch algorithms. Moreover, it exhibits good tracking capability for multiple targets with varying location parameters.