Space-time Adaptive Processing via Fast Environment Sensing

Abstract

In the heterogeneous clutter environment, the traditional space-time adaptive processing (STAP) cannot accurately estimate the clutter covariance matrix (CCM) due to the lack of training samples, so its clutter suppression performance is seriously degraded. The STAP based on dynamic environment perception can obtain the accurate estimation of CCM under the condition of single training sample, which greatly improves the ability of STAP to suppress heterogeneous clutter, unfortunately, this method suffers from low computational efficiency. This paper proposes a STAP via fast environment sensing algorithm to solve the problem. This algorithm first estimates the number of strong clutter patches in the clutter scene by beam scanning and uses it as the iterative stopping condition of the OMP algorithm. Then the clutter scene is reconstructed using the OMP algorithm as clutter prior information. Finally, the clutter-plus-noise covariance matrix (CNCM) is constructed using clutter prior information for STAP. The simulation results show that, compared with the existing environment sensing algorithm, the method proposed in this paper greatly improves computational efficiency and simultaneously has exhilarant clutter suppression performance.