MPSK Signal Modulation Recognition based on Decision Fusion

Abstract

Many scholars have done a lot of research on the modulation recognition of signals, and have proposed various modulation recognition methods. However, in the commonly used recognition methods, the recognition effect of the multiple phase shift keying signal (Multiple Phase Shift Keying, MPSK) is not ideal. Therefore, a decision fusion method based on a deep learning model is proposed. Firstly, I/Q modulation is used to obtain the MPSK modulated signal, and then through signal preprocessing, the time-frequency diagram and constellation diagram of the modulated signal are received. Next, the feature extraction and recognition are performed through the convolutional neural network, and finally, the results of the two types of feature recognition are combined with DS evidence theory to obtain the final recognition result. It can be seen from the simulation experiment that under different signal-to-noise ratios, the recognition effect of the time-frequency diagram and the constellation diagram has its own advantages, and each advantage can be used to make decision fusion, strengthen the generalization ability of the recognition model, and make the recognition effect of MPSK signal better. And the recognition accuracy of decision fusion is higher than 90%, when at the high signal-to-noise ratio.