Intensive Interrupted Sampling Repeater Jamming Detection Based on Transformer-CFAR Fusion Detection Model

Abstract

Interrupted sampling repeater jamming (ISRJ) is an effective way to suppress normal radar probing in modern electronic warfare, and ISRJ suppression is essential in the radar signal processing stage. Accurately locating the position of the jamming signal is the first step to suppress the jamming signal. However, the traditional constant false alarm rate (CFAR) has performance loss with the characteristics of intensive distribution, extensive dynamic power range, and nonperiodical transmitting. This article proposes an ISRJ detection method based on a fusion Transformer and CFAR. First, the feature extraction model based on the Transformer is built to extract the continuous distribution features of ISRJ, forming rough detection results. Then, a CFAR detector is derived based on the rough detection result to estimate detector parameters. Third, the jamming features obtained by the Transformer, CFAR detector, and other features (such as instantaneous power and average power) are fused by a decision tree to realize robust detection in a low jamming-to-signal noise ratio (JSNR). Finally, we conducted simulated experiments to verify the effectiveness of the proposed method.