Communication radiation source individual identification method based on double-path attention convolutional neural network

Abstract

The invention discloses a communication radiation source individual identification method based on a double-path attention convolutional neural network, and relates to the technical field of radio frequency fingerprint identification. According to the communication radiation source individual identification method based on the double-path attention convolutional neural network provided by the invention, radio frequency fingerprint features are automatically extracted by directly utilizing radio frequency IQ signals collected by a receiver; the importance of each dimension in the radio frequency fingerprint feature vector is analyzed through a space attention branch in the double-path attention convolutional neural network, and the respective importance of various radio frequency fingerprint features is automatically judged through a channel attention branch; according to the method, multiple radio frequency fingerprint features are obtained, the multiple radio frequency fingerprint features are fused in the channel dimension and the space dimension at the same time according to importance, radiation source radio frequency fingerprint features with high discrimination capability are obtained, and therefore the individual identification efficiency of different types of radiation sources is improved, and the accuracy of individual identification tasks is improved.

Description

technical field [0001] The invention relates to the technical field of radio frequency fingerprint identification, in particular to a communication radiation source individual identification method based on a two-way attention convolutional neural network. Background technique [0002] As one of the common key research topics in the fields of wireless spectrum monitoring and Internet of Things security communication, the individual identification of communication radiation sources refers to the detection of transmission signals of communication radiation sources, and the use of signal processing, machine learning, pattern recognition and other technologies to extract information that can be distinguished from them. The radio frequency fingerprint characteristics of different communication radiation sources can be used to distinguish individual communication radiation sources through classification or clustering algorithms to realize individual identification. [0003] The ex...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above-mentioned problems existing in the prior art, the present invention provides a communication radiation source individual identification method based on a two-way attention convolutional neural network, by automatically extracting a variety of radio frequency fingerprint features in a data-driven manner, and automatically judging different features The importance and feature fusion are carried out, so as to solve the technical problems of low efficiency and low accuracy in the identification of different types of individual radiation sources in the existing individual identification methods of communication radiation sources, so as to improve the efficiency and accuracy of individual identification of different types of radiation sources technical effect

Images