Automatic Modulation and Recognition of Robot Communication Signal Based on Deep Learning Neural Network

Abstract

In order to solve the problem that the traditional method of manually extracting expert features for communication signal recognition has large limitations and low accuracy under low signal-to-noise ratio, this paper proposes an automatic modulation and recognition method of robot communication signal based on deep learning neural network. In this method, the received signal is preprocessed to obtain the complex baseband signal including in-phase component and quadrature component. The signal is used as the data set of the input convolution neural network model. The model structure and the super parameters such as convolution kernel, step size, characteristic graph, and activation function are adjusted through multiple training, and the trained model is used to extract and recognize the features of the communication signal. It realizes the identification and classification of seven types of digital communication signals: 2FSK, 4FSK, BPSK, 8PSK, QPSK, QAM16, and QAM64. The experimental results show that the average recognition accuracy of the seven signals has reached 94.61% when the signal-to-noise ratio is 0 dB. Conclusion. The algorithm is proved to be effective and has high accuracy under the condition of low signal-to-noise ratio.