Learning of Time-Frequency Attention Mechanism for Automatic Modulation Recognition

Recently, deep learning-based image classification and speech recognition approaches have made extensive use of attention mechanisms to achieve state-of-the-art recognition, which demonstrates the effectiveness mechanisms. Motivated by fact that frequency time information modulated radio signals are crucial for modulation this letter proposes a time-frequency mechanism convolutional neural network (CNN)-based automatic recognition. The proposed is designed learn channel, more meaningful in CNN We analyze evaluate performance models. Experiment results show methods outperform existing