Fine-Grained Recognition and Suppression of ISRJ Based on UNet-A

Interrupted-sampling and repeater jamming (ISRJ), as a novel form of active jamming, has emerged as a focal point and challenge in radar jamming countermeasures. In this letter, to enhance the suppression capability against ISRJ, we propose a recognition and suppression method based on the UNet-attention (UNet-A) semantic segmentation model. First, an attention-based direct connection structure between the encoder and decoder is designed to enhance the ability of UNet-A to identify the boundaries of jamming and the target. Then, an adaptive time-frequency (TF) filter based on the refined recognition results is designed to improve the signal-to-jamming ratio improvement factor (SJRIF). Finally, to improve the jamming suppression capability while reducing the target energy loss, an annotation method based on the target energy loss constraint criterion is proposed, and a dataset is constructed based on this. Numerical results and comparisons with the existing methods are included to demonstrate that the proposed method can effectively enhance anti-ISRJ performance.