Compound Jamming Recognition Under Low JNR Setting Based on a Dual-Branch Residual Fusion Network.

Area of Science:

• Electrical Engineering
• Computer Science
• Signal Processing

Background:

• Radar systems face significant challenges from advanced jamming techniques in complex electromagnetic environments.
• Existing recognition methods struggle with the diversity, complexity, and low signal-to-noise ratio conditions of modern jamming signals.

Purpose of the Study:

• To develop a novel deep learning approach for accurate and robust recognition of complex radar jamming signals.
• To improve radar anti-jamming capabilities against sophisticated threats.

Main Methods:

• A dual-branch deep learning network processing both time-domain and time-frequency-domain features.
• Incorporation of a multi-branch convolutional structure for enhanced feature extraction.
• Application of an effective feature fusion strategy to capture subtle jamming patterns.

Main Results:

• The proposed deep learning method demonstrated superior recognition accuracy compared to six baseline approaches.
• The approach showed significant robustness, particularly under low jamming-to-noise ratio conditions.
• Outperformed existing methods in identifying diverse and complex jamming signals.

Conclusions:

• The novel deep learning approach offers a promising solution for complex radar jamming signal recognition.
• The dual-branch architecture and feature fusion strategy effectively enhance anti-jamming performance.
• This method is critical for advancing radar system resilience in contested electromagnetic environments.