Feature selection enhanced sparse recovery space-time adaptive processing

This paper presents a novel radar moving target detection method that integrates feature selection-enhanced sparse recovery within a space-time adaptive processing (STAP) framework. The proposed method addresses the performance limitation of airborne radar systems due to large data volume and high computational complexity. By pre-screening and weighting the dictionary atoms through the feature selection mechanism, the method not only reduces the effective sampling rate, but also improves the reconstruction accuracy of the target signal. Simulation results show that the method can significantly suppress clutter in complex and inhomogeneous environments, thus improving detection accuracy and efficiency. The proposed scheme has a broad application prospect in the fields of national security and disaster relief.