JSR-GFNet: Jamming-to-Signal Ratio-Aware Dynamic Gating for Interference Classification in future Cognitive Global Navigation Satellite Systems

TL;DR

This work addresses GNSS interference classification under dynamic jamming by proposing JSR-GFNet, a physics-aware multi-modal network that fuses phase-sensitive IQ data and STFT-based spectrograms through a decoupled dynamic gating mechanism guided by JSR-derived statistics. It introduces the CGI-21 dataset with 21 interference classes and extensive JSR variation to benchmark performance, showing superior accuracy across $JSR$ in the $10$ to $50$ dB range and providing interpretability through gating behavior that mirrors signal-processing intuition. The approach resolves phase-degeneracy issues inherent to magnitude-only spectrograms and demonstrates favorable computational efficiency, making it suitable for near-real-time cognitive anti-jamming in next-generation aerospace GNSS receivers. Overall, JSR-GFNet advances robust GNSS interference classification by leveraging complementary modalities and adaptive fusion, with practical implications for security and resilience of critical navigation infrastructure.

Abstract

The transition toward cognitive global navigation satellite system (GNSS) receivers requires accurate interference classification to trigger adaptive mitigation strategies. However, conventional methods relying on Time-Frequency Analysis (TFA) and Convolutional Neural Networks (CNNs) face two fundamental limitations: severe performance degradation in low Jamming-to-Signal Ratio (JSR) regimes due to noise obscuration, and ``feature degeneracy'' caused by the loss of phase information in magnitude-only spectrograms. Consequently, spectrally similar signals -- such as high-order Quadrature Amplitude Modulation versus Band-Limited Gaussian Noise -- become indistinguishable. To overcome these challenges, this paper proposes the \textbf{JSR-Guided Fusion Network (JSR-GFNet)}. This multi-modal architecture combines phase-sensitive complex In-Phase/Quadrature (IQ) samples with Short-Time Fourier Transform (STFT) spectrograms. Central to this framework is a physics-inspired dynamic gating mechanism driven by statistical signal descriptors. Acting as a conditional controller, it autonomously estimates signal reliability to dynamically reweight the contributions of a Complex-Valued ResNet (IQ stream) and an EfficientNet backbone (STFT stream). To validate the model, we introduce the Comprehensive GNSS Interference (CGI-21) dataset, simulating 21 jamming categories including software-defined waveforms from aerial platforms. Extensive experiments demonstrate that JSR-GFNet achieves higher accuracy across the full 10--50 dB JSR spectrum. Notably, interpretability analysis confirms that the model learns a physically intuitive strategy: prioritizing spectral energy integration in noise-limited regimes while shifting focus to phase precision in high-SNR scenarios to resolve modulation ambiguities. This framework provides a robust solution for next-generation aerospace navigation security.

Figures (9)

• Figure 1: Schematic representation of the proposed JSR-GFNet architecture.
• Figure 2: Spectrogram representations of the simulated jamming signals generated by STFT.
• Figure 3: Detailed architecture of the EfficientNet-B0 backbone, featuring the stack of MBConv blocks and the final classification head.
• Figure 4: Illustration of the terrestrial interference monitoring scenario involving an Aerial Autonomous Vehicle (AAV) threat.
• Figure 5: Overall classification accuracy comparison of JSR-GFNet against single-modality baselines.