Robust HRRP Recognition under Interrupted Sampling Repeater Jamming using a Prior Jamming Information-Guided Network

TL;DR

The paper tackles the challenge of radar automatic target recognition from HRRPs under interrupted-sampling repeater jamming (ISRJ), where distortions severely degrade features. It introduces a prior jamming information representation via a point spread function (PSF) and embeds this prior into a neural network through a prior-guided feature interaction module and a hybrid loss that combines cross-entropy with supervised contrastive learning. The key contributions are the PSF-based ISRJ distortion modeling, a dual-attention fusion and prior-guided feature selection mechanism, and a loss design that improves discriminability under distribution shifts. Experiments on simulated and measured data show state-of-the-art robustness and strong generalization to unseen jamming parameters, indicating practical potential for ECM-resilient HRRP recognition.

Abstract

Radar automatic target recognition (RATR) based on high-resolution range profile (HRRP) has attracted increasing attention due to its ability to capture fine-grained structural features. However, recognizing targets under electronic countermeasures (ECM), especially the mainstream interrupted-sampling repeater jamming (ISRJ), remains a significant challenge, as HRRPs often suffer from serious feature distortion. To address this, we propose a robust HRRP recognition method guided by prior jamming information. Specifically, we introduce a point spread function (PSF) as prior information to model the HRRP distortion induced by ISRJ. Based on this, we design a recognition network that leverages this prior through a prior-guided feature interaction module and a hybrid loss function to enhance the model's discriminative capability. With the aid of prior information, the model can learn invariant features within distorted HRRP under different jamming parameters. Both the simulated and measured-data experiments demonstrate that our method consistently outperforms state-of-the-art approaches and exhibits stronger generalization capabilities when facing unseen jamming parameters.

Figures (8)

• Figure 1: Time–frequency representations and corresponding HRRPs of radar echoes under different conditions.
• Figure 2: Overall framework of the proposed HRRP recognition network.
• Figure 3: Process of generating HRRP and prior jamming information.
• Figure 4: Visualization of HRRP and the corresponding prior jamming information under ISRJ with different jamming configurations.
• Figure 5: Visualization of feature distributions learned through supervised contrastive learning.