On the Exact Success Rate of Side Channel Analysis in the Gaussian Model

Nowadays, Side Channel Analysis is one of the most powerful cryptanalytic technique against cryptosystems embedded in portable devices such as smart cards. Faced with this threat, it is of crucial importance to precisely determine what is achievable by a given side channel adversary against a cryptosystem producing a given side channel leakage. This can be answered by evaluating the success rate of an attack according to the adversary capacities and to the leakage properties. In this paper, we investigate the issue of evaluating the success rate of side channel analysis in the widely admitted Gaussian leakage model. We introduce a new approach that allows us to efficiently compute the success rate of an attack in this model and we apply it to the two main families of side channel analysis: differential side channel analysis and profiling side channel analysis.