A Novel Approach for Automatic Recognition of LPI Radar Waveforms Based on CNN and Attention Mechanisms

Abstract

Low probability of intercept (LPI) radar waveform identification is a challenging task in modern radar and electronic warfare (EW) systems. A high precision LPI radar waveform automatic recognition method based on CNN and attention mechanism is proposed in this paper. Firstly, using the smoothed pseudo-Wigner-Ville distribution (SPWVD) to convert the reconnaissance signal to a time-frequency image. Then the time-frequency images are preprocessed by using bicubic interpolation. Finally, the automatic recognition of the LPI radar waveform is realized by LRCNet which fused the convolutional neural network (CNN) and convolutional block attention mechanism (CBAM). The results show that when the signal-to-noise ratio (SNR) is −4 dB, the model achieved an overall average recognition rate of 99.2% for 13 LPI radar waveforms (LFM, Rect, Costas, Barker, Fank, P1-P4, T1-T4 codes). Compared with other typical waveform recognition methods, the proposed method has an obvious improvement in recognition accuracy under low SNR.