Paper 2024/049

CL-SCA: Leveraging Contrastive Learning for Profiled Side-Channel Analysis

Abstract

Side-channel analysis based on machine learning, especially neural networks, has gained significant attention in recent years. However, many existing methods still suffer from certain limitations. Despite the inherent capability of neural networks to extract features, there remains a risk of extracting irrelevant information. The heavy reliance on profiled traces makes it challenging to adapt to remote attack scenarios with limited profiled traces. Besides, attack traces also contain critical information that can be used in the training process to assist model learning. In this paper, we propose a side-channel analysis approach based on contrastive learning named CL-SCA to address these issues. We also leverage a stochastic data augmentation technique to assist model to effectively filter out irrelevant information from the profiled traces. Through experiments of different datasets from different platforms, we demonstrate that CL-SCA significantly outperforms various conventional machine learning side-channel analysis techniques. Moreover, by incorporating attack traces into the training process using our approach, known as CL-SCA+, it becomes possible to achieve even greater enhancements. This extension can further improve the effectiveness of key recovery, which is fully verified through experiments on different datasets.