基于卷积神经网络的数字调制分类识别

Classification and recognition of digital modulation based on Convolutional neural network

基于特征提取和模式识别的信号调制方式分类识别技术是非协作通信领域内应用广泛的重点研究对象。提出一种基于深度学习的通信信号数字调制识别算法,采用卷积神经网络找到数据的内在表达,实现逐层化地识别和分类MPSK、MFSK和MQAM中的六种调制信号。仿真实验结果表明,所提出的方法分类识别效果良好,基本达成了数字调制信号自动识别的目的。

The signal modulation classification and recognition technology based on feature extrac⁃tion and pattern recognition is a widely used key research object in the field of non cooperative com⁃munication.A digital modulation recognition algorithm for communication signals based on deep learning is proposed.The Convolutional neural network is used to find the internal expression of data,and the six modulation signals in MPSK,MFSK and MQAM are recognized and classified layer by layer.The simulation experiment results show that the proposed method has good classification and recognition performance,and basically achieves the goal of automatic recognition of digital modulation signals.