Electromagnetic Information Leakage Classification Detection Method for Computer Display Equipment Based on Machine Learning

Abstract

The detection of electromagnetic information leakage is aimed at confirming security vulnerabilities that existing in electronic devices. Computer display is a kind of important electronic device. The traditional method of detecting electromagnetic information leakage from computer displays is to reconstruct original image. It is necessary for reconstruction technology to understand the working mechanism of the detected equipment and other relevant information. However, this process tends to be complex so that it is hard to realize the detection. This paper proposes a classification detection method based on Machine Learning. It can detect electromagnetic information leakage by classifying and identifying the electromagnetic information radiated from computer display equipment. In experiment, we preprocess electromagnetic signal through Wavelet Transform, design a three-layer convolutional neural network, classify the samples correctly. The detection method based on Machine Learning overcomes the technical drawback that traditional detection methods need to reconstruct. We compare with other common Machine Learning algorithms. The experiment result shows that the method based on three-layer convolutional neural network and wavelet transform has good accuracy in the classification and detection of electromagnetic leakage of computer monitors.