Modulation signal identification method based on high-order cumulant and high-order cumulant after difference

• Main Authors: HUO YUHANG, LI PEIJIAN, ZHANG YU, CHU XIAOPENG, LI XIANG, GONG ZHENG
• Format: Patent
• Language: Chinese; English
• Published: 10.03.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PHYSICS; TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHICCOMMUNICATION

The invention provides a modulation signal identification method based on a high-order cumulant and a high-order cumulant after difference, and the method comprises the steps: carrying out the preprocessing of a collected modulation signal, carrying out the feature parameter extraction of a processed result, taking the extracted feature parameter as the input of a neural network model, and carrying out the recognition of the high-order cumulant after the feature parameter passes through the neural network model. And outputting a final signal modulation type discrimination result. According to the method, high-order cumulant and differentiated high-order cumulant features are adopted as signal features for neural network training, a trained neural network model serves as a classification recognizer for signal classification recognition, and nine kinds of signals including 2PSK, QPSK, OQPSK, 8PSK, 16PSK, Pi-4QPSK, 8QAM, 16QAM and 32QAM can be distinguished. And modulation identification of more types of signals