Intra-pulse modulation recognition of radar signals based on multi-feature random matching fusion network

Intra-pulse modulation recognition of radar signals plays an important role in the field of electronic warfare. In this paper, a multi-feature random matching fusion (MFRMF) network is proposed to deal with the recognition technology of radar signals' intra-pulse modulation at a low signal-to-noise ratio (SNR). First, we extract 12 traditional parameter features of radar signals and screen out 7 more important features. Next, we analyze and extract the Time-frequency images. Finally, the MFRMF network with the idea of residual learning, self-attention mechanism, and random matching algorithm is adopted to perform feature learning and identify the intra-pulse modulation type of radar signals. Simulation results demonstrate that MFRMF can effectively reduce the interference of noise on signal classification and improve recognition accuracy at a low SNR. It can classify 10 kinds of radar signals, and the overall recognition accuracy achieves 90.6% and 95.4% when the SNR is - 8 dB and - 6 dB, respectively.