Automatic Modulation Recognition for Radar Signals Based on ACSE Networks
Qizhe Qu, Yongliang Wang, Qinglei Du · 2021 CIE International Conference on Radar (Radar) · 2021
Automatic modulation recognition for radar signals plays a significant role in electronic warfare. Conventional recognition methods may suffer from the recognition accuracy and the computation complexity under low signal-to-noise ratio (SNR). In this paper, a novel method based on Asymmetric Convolution Squeeze-and-Excitation (ACSE) networks and a fusion strategy is proposed to recognize seven kinds of radar signals automatically. First, features in these three domains are converted into images. Then the images are as inputs for ACSE networks which can extract and learn deep features without complex data pre-processing. Finally, inference results in the three domains are fused using voting strategy of ensemble learning to output the final inference result. Via simulations, the robustness and effectiveness of voting strategy are proved. The results on simulation signals show that the proposed method can achieve more than 95% accuracy at −6dB. Compared with SVM method and three typical neural networks, the ACSE networks owns better recognition performance especially under low SNRs. The results on measured signals also demonstrate that the proposed method outperforms other methods.