Research on Side-Channel Attack Based on the Synergy between SNR and Convolutional Neural Networks

Abstract

The data encryption process in the encryption chip will be leaked by means of timing operation time, probing data collecting operation power and electromagnetic signal interception, which makes side channel attack possible. Nowadays, the research on template creation of template attacks has shifted from Gaussian distribution to the use of machine learning algorithms to create templates. With many parameters, it is difficult to find a suitable network structure. Based on many experimental studies, the experience of a convolutional neural network structure suitable for side channel analysis is summarized and proposed.