Automatic Modulation Classification Based on CNN, LSTM and Attention Mechanism

Automatic modulation recognition (AMR) is a key technology in the field of cognitive radio. To improve the performance, a new model for AMR, which incorporates an attention mechanism to extract the important features of the signal more effectively, is proposed in this paper. The simulation results show that our proposed method has a good recognition performance. In addition, our method can still achieve high accuracy, even if only a few samples are provided as the training set for training. We also visualize the output features of the model and show the positive effect of the attention mechanism on signal modulation category identification.