Open Set Recognition of Communication Jamming Patterns with Raw I/Q Data
Ziming Du, Jiahao Li, Bo Zhou, Wei Wang, Qihui Wu · 2024
Effective recognition of communication jammings is essential to maintain the integrity of the electromagnetic spectrum space. In this paper, we propose a novel open-set jamming pattern recognition method by directly utilizing raw I/Q data. First, we design an I/Q data feature enhancement module to capture the interaction features between the I/Q channels. Then, we integrate the raw I/Q data with their interaction features to extract jamming characteristics for known patterns. Finally, we develop a distance-based open-set classifier to recognize both known and unknown patterns. Simulation results show that our proposed method not only achieves superior recognition accuracy but also exhibits strong robustness when the training and test datasets differ in channel conditions and jamming-to-signal ratios.