Radar signal modulation mode identification method based on attention mechanism

According to the radar signal modulation mode identification method based on the attention mechanism, original data signal features of signals are extracted through a long-short term memory network, weights are distributed to learned features according to the attention mechanism, and finally, a classifier outputs a classification result according to the learned features. And performing code implementation on the model by utilizing a PyTorch framework of the Python language. A simulation experiment is carried out on a common radar signal system, it can be seen that the algorithm is high in prediction accuracy and good in stability, and it is indicated that the model has feasibility and effectiveness in recognition of common radar signal modulation modes. 基于注意力机制的雷达信号调制方式识别方法，通过长短时记忆网络提取信号原始数据信号特征，根据注意力机制为学习到的特征分配权重，最后，分类器根据学习到的特征输出分类结果。利用Python语言的PyTorch框架对模型进行代码实现。通过对常见信号雷达信号体制进行过仿真实验，可以看出该算法的预测准确率较高，稳定性好，表明了此模型对常见雷达信号调制方式识别是具有可行性和有效性的。