Research on Combined Interference Suppression Technique of UBSS based on Compressed Sensing

In radar communications, the complex combined interference severely limits system performance, and traditional suppression techniques have certain limitations. This paper focuses on the combined interference suppression technology based on compressed sensing. It first analyzes the sparse transformation methods of signals, which are beneficial for processing non - stationary signals and extracting local features. For the underdetermined blind source separation, a time - frequency independent complex angle detection method is used to estimate the mixing matrix. Specifically, by detecting the complex angle in the time - frequency domain of radar signals, the column vectors of the mixing matrix can be obtained. In terms of signal recovery, a method based on compressed sensing and improved l1 - norm minimization is proposed. This method links underdetermined blind source separation with the compressed sensing model and applies a greedy algorithm to approximate the optimal solution for signal recovery. The simulation results show that, compared with the traditional algorithm, the proposed algorithm has better performance in separating target signals from interference, especially in low-signal-to-noise ratio environments. It can effectively improve the signal processing ability of radar systems in complex interference environments.