Clutter Suppression Algorithm via Covariance Matrix Reconstruction With Airborne Passive Radar

Clutter Suppression is a crucial problem in Airborne Passive Radar (APR) systems. However, some classic Space-Time Adaptive Processing (STAP) algorithms suitable for Airborne Active Radar (AAR) cannot be directly ported to the APR due to the continuity, randomness and non-periodicity of the illuminator of opportunity in APR. By analyzing the differences between AAR systems and APR systems and taking into account the characteristics of illuminator of opportunity, a special segmentation operation suitable for passive radar and a Robust Two-Dimensional Atomic Norm Minimization (R2D-ANM) model are developed in this paper for accurately estimating the Clutter Covariance Matrix (CCM). From the constructed model, an equivalent SemiDefinite Program (SDP) is derived and the regularization factor about the convex optimization problem is also theoretically obtained. In addition, this paper provides the optimization conditions and recovery performance bound for the constructed optimization problem. To avoid the heavy computational load and large dynamic storage, an expression with Alternating Direction Method of Multipliers (ADMM) is presented. Finally, a series of experiments demonstrate that the STAP based on R2D-ANM (R2D-ANM-STAP) algorithm outperforms the existing algorithms for APR systems.