A New Radar Signal Modulation Recognition Algorithm Based on Time-frequency Transform

To solve the problem that exist methods of radar signal modulation recognition do not perform well under low SNR, we propose a new modulation recognition algorithm based on time-frequency transform. The algorithm applies time-frequency transform to analysis radar signal in time-frequency domain. Meanwhile, AlexNet, Hu moments and t-SNE are the detailed methods in algorithm to improve the recognition rate. The AlexNet can extract the detailed feature of radar signal based on transfer learning, while the Hu moments can express the shape feature. Thus the fusion of two kinds of feature further improve the recognition rate. The simulation results show that the proposed algorithm can deal with the small sample of radar signal under low SNR. And the recognition rate is 95% at the SNR of 4dB. The proposed algorithm can be employed in electronic reconnaissance for effective reconnaissance under complex environment.