The use of ellipse-based estimator as a sub-key distinguisher for Side-Channel Analysis. (September 2021)

• Record Type: Journal Article
• Authors: Martínez-Herrera, Alberto F.; Mex-Perera, Carlos; Mancillas-López, Cuauhtemoc; Del-Valle-Soto, Carolina; Bossuet, Lilian
• Is Part Of: Computers & electrical engineering. Volume 94(2021)
• Publication Date: 2021-09
• Subjects: Side-Channel Analysis -- Ellipse-based estimator -- Power traces -- Success rate -- Sub-keys; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2021.107311
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Side-Channel Analysis has become a relevant tool to analyze a cryptographic device. Here, an adversary looks for information leakage from emanation sources such as power consumption, thus obtaining sensitive information with a lower effort than the mathematical approach. In this manuscript, a distinguisher based on the Mahalanobis distance is applied. Instead of computing an inverse covariance matrix because of lack-sampling problems with tested datasets, a shrinkage calculation is implemented, thus obtaining efficient Mahalanobis distance implementations. The current approach is evaluated using different standardized tests such as stability and success probability, which are computed using unmasked public traces obtained from the Advanced Encryption Standard's typical implementations with a 128-bit key. We show that this technique's efficiency is better than Pearson correlation using few traces in terms of probability detection. Graphical abstract: Highlights: The Mahalanobis distance is represented with equivalent identities for computation efficiency. These efficient representations helps to mount a reliable Side Channel distinguisher. It is feasible to deal with low-sampled traces with the use of shrinkage factors. The shrinkage factors do not change the computation reliability of this distinguisher. These efficient representations do not lose precision in the gathering of the private key. Mostly, these representations have a better detection rate than the correlation-based approach.