Using principal component analysis for practical biasing of power traces to improve power analysis attacks

Abstract

Researchers have focused significant attention on sidechannel attacks since the first power analysis attack was introduced. To date, several ideas have been introduced to efficiently analyze cryptographic modules. A power trace selection method for improving attack efficiency was recently presented; however, applying it involves many restrictions. Therefore, we propose a new selection method to improve power analysis attacks using principal component analysis. Our method is a practical one for biasing power traces. Our experimental results show that the proposed method improves attack efficiency in terms of the number of traces used for finding the secret key.