Deep Neural Network Attribution Methods for Leakage Analysis and Symmetric Key Recovery

摘要

Deep Neural Networks (DNNs) have recently received significant attention in the side-channel community due to their state-of-the-art performance in security testing of embedded systems. However, research on the subject mostly focused on techniques to improve the attack efficiency in terms of the number of traces required to extract secret parameters. What has not been investigated in detail is a constructive approach of DNNs as a tool to evaluate and improve the effectiveness of countermeasures against side-channel attacks. In this work, we close this gap by applying attribution methods that aim for interpreting Deep Neural Network (DNN) decisions in order to identify leaking operations in cryptographic implementations. In particular, we investigate three different approaches that have been proposed for feature visualization in image classification tasks and compare them regarding their suitability to reveal Points of Interest (POIs) in side-channel traces. We show by experiments with four separate data sets that the three methods are especially interesting in the context of side-channel protected implementations and misaligned measurements. Finally, we demonstrate that attribution can also serve as a powerful side-channel distinguisher leading to a successful retrieval of the secret key with at least five times fewer traces compared to standard key recovery in DNN-based attack setups.