Frequency hopping signal modulation identification method based on graph convolutional neural network

The invention provides a frequency hopping signal modulation identification method based on a graph convolutional neural network. The method comprises the following steps of: 1, extracting node characteristics of a frequency hopping signal; 2, constructing an adjacent matrix and an edge, and converting the frequency hopping signal into an undirected topological graph; 3, constructing a graph convolutional neural network GCN model according to the frequency hopping signal graph domain conversion data; and 4, training the GCN model by using the training sample set, inputting the test sample into the trained GCN model, and outputting an identification result. According to the signal graph domain conversion method provided by the invention, on the basis of reducing the number of nodes and edges, various node features are extracted, the parameters and the calculation amount are reduced, the anti-noise performance is good, the constructed GCN model can obtain space structure information which ...