Electromagnetic signal recognition method and device for constructing graph convolutional network based on implicit knowledge

Abstract

The invention discloses an electromagnetic signal recognition method and device for constructing a graph convolution network based on implicit knowledge, wherein the method includes: acquiring a plurality of electromagnetic signals, and extracting feature descriptions of the plurality of electromagnetic signals, and according to the plurality of electromagnetic signals The feature description of the generated feature matrix is ​​used as the first type of input information; the implicit knowledge of multiple electromagnetic signals is mined, and the graph structure based on the implicit knowledge of electromagnetic signals is constructed according to the implicit knowledge as the second type of input information; according to the first type of input information Build a graph convolutional neural network with the second type of input information, and use the weakly supervised learning method to iteratively train the graph convolutional neural network, so that the graph convolutional neural network can output multiple categories of electromagnetic signals. This method can use the characteristics of each node, mine the relationship between nodes, extract the characteristics of each electromagnetic signal category in depth, construct a graph convolutional neural network, identify electromagnetic signals, and use more information , with stronger generalization ability.

Description

technical field [0001] The invention relates to the technical field of electromagnetic signal intelligent perception, in particular to an electromagnetic signal recognition method and device for constructing a graph convolution network based on implicit knowledge. Background technique [0002] Nowadays, with the continuous popularization of mobile communication devices and the vigorous development of Internet of Things technology, especially in the context of the large-scale application of 5G technology and Internet of Things technology, the protection of wireless communication security such as device verification, illegal emission device monitoring, etc., Whether it is in civil communication or industrial manufacturing, it has become more and more important. The extraction of signal subtle features has shown great application value in this field. It can use signal processing technology to identify wireless communication signals and extract subtle features, and then cooperat...

AI Technical Summary

Problems solved by technology

However, it should be noted that these identification methods relying on artificial design are difficult to guarantee the accuracy of target identification and the response speed of identification in the current increasingly complex electromagnetic environment.

Images