Optimizing Inner Product Masking Scheme by a Coding Theory Approach

Masking is one of the most popular countermeasures to protect cryptographic implementations against side-channel analysis since it provably secure and can be deployed at algorithm level. To strengthen original Boolean masking scheme, several works have suggested using schemes with high algebraic complexity. The Inner Product (IPM) those. In this paper, we propose a unified framework quantitatively assess security IPM in coding-theoretic approach. Specifically, starting from expression coded form, use two defining parameters code characterize its resistance. order validate framework, then connect leakage metrics (namely signal-to-noise ratio mutual information, an information-theoretic aspect) typical attack metric (success rate, practical build firm foundation for our framework. As application, results provide ultimate explanations on observations made by Balasch et al. EUROCRYPT'15 ASIACRYPT'17, Wang CARDIS'16 Poussier CARDIS'17 regarding parameter effects IPM, like higher bounded moment model. Furthermore, show how systematically choose optimal codes (in sense concrete level) optimize Eventually, present simple but effective choosing which special interest designers when selecting IPM.