Time-Frequency Fusion Deep Learning Algorithm for Modulation Recognition

This study explores the use of deep learning for signal modulation recognition in complex electronic warfare environments. By extracting time-frequency features from 12 types of radar and communication signals using Wigner-Ville Distribution and Wavelet Transform, the method provides refined signal representations. Leveraging deep learning for feature classification, it achieves accurate modulation recognition, effectively addressing noise and interference challenges. The approach enhances recognition accuracy and robustness, offering a novel solution for radar signal modulation identification in modern electronic warfare systems.