Emitter Recognition Method Based on Feature Fusion

In order to solve the problem that the fingerprint features of communication emitter extracted based on the traditional mechanism model are not detailed enough, which leads to the low identification rate of individual communication emitter, a method for individual identification of communication emitter based on feature fusion is proposed. The method extracts the fingerprint features based on traditional mechanism models, such as statistical features of the signal, and then constructs a deep neural network to extract deep and subtle features from the original signal. The extracted multi-dimensional features are combined, and the fused features are used to identify the individual communication emitter through the random forest classification algorithm. The research results show that compared with the emitter individual identification method based on the traditional mechanism model, the feature fusion method proposed in this paper makes full use of the traditional fingerprint features of the communication emitter signal, and combines the powerful subtle feature extraction ability of the neural network to achieve In order to better identify the individual communication emitter under the condition of low signal-to-noise ratio, the experimental results show that this method can significantly improve the accuracy of individual emitter identification.