Interrupted sampling repeater jamming suppression based on Gaussian mixture model and sparse reconstruction

Abstract

The interrupted sampling repeater jamming (ISRJ) based on digital radio frequency memory (DRFM) is coherent with the radar transmitted signal, allowing it to obtain partial pulse compression gain while exhibiting both suppression and deception jamming effects, posing a significant threat to modern radars. In this article, we propose an ISRJ suppression method based on Gaussian Mixture Model (GMM) and sparse reconstruction. Firstly, the probability distribution of the amplitudes of the received echoes is fitted using a GMM to obtain the mean value of each component. Based on the magnitude and the discontinuous nature of ISRJ in the time domain, jamming-free signal segments are extracted. Subsequently, these segments are processed through de-chirping, and by leveraging their sparse characteristics in the frequency domain, the complete target echo is obtained through sparse reconstruction. Matching filtering is then applied to the reconstructed target echo, followed by peak searching, to detect the genuine target echo, thereby achieving ISRJ suppression. Experimental results demonstrate that this method exhibits high accuracy in extracting the jamming-free signal segments and excellent ISRJ suppression performance, while also demonstrating strong robustness against different jamming parameters.