Fuzzy analysis of side channel information
Ehsan Saeedi, Yinan Kong · 2014
Cryptographic algorithms implemented inside embedded systems have been designed to guarantee the security of data transition, but these electronic devices can be vulnerable to different class of attacks, mostly side channel attacks. Over the last decade, many attacks have been done successfully on different cryptographic algorithms, but these attacks mostly benefit from different ideal assumption and laboratory conditions. In this paper we propose a novel approach of side channel information analysis, based on fuzzy logic, that can efficiently conjecture the secret bits of a cryptosystem especially in realtime condition in which inputs are usually imprecise, noisy or missing. The validation and effictiveness of the proposed fuzzy approach is shown through a case study and accuracy comparison between electromagnetic, power and fuzzy analysis methods. Our experimental results confirmed that the fuzzy analysis produce comparable results, respectively better in the presence of noise.