A novel STAP method based on structured sparse recovery of clutter spectrum

In this paper we propose a novel STAP method based on the structured sparse recovery of radar clutter spectrum. Besides the spatial-temporal sparsity, we introduce the structured property of the clutter spectrum in STAP based on the pattern of 2-dimensional clutter spectrum. An elliptical clustering model is given to describe the structured sparsity, based on which a novel sparse recovery STAP method named SSR-STAP is developed. In this new method, we model the clutter structured property as a priori based on Markov random field. And we propose an improved FOCUSS algorithm named Elliptical Clustering FOCUSS by introducing a priori information of clutter spectrum structure into an iterative Bayesian estimation process of weight coefficients. The simulation results show that the performances of SSR-STAP method are superior to the previous Sparse Recovery based Space-Time Adaptive Processing (SR-STAP) method both in clutter suppression and moving target detection.