A model of DPA syndrome and its application to the identification of leaking gates

Abstract

Within the context of secure applications, side channel attacks are a major threat. The main characteristic of these attacks is that they exploit physical syndromes, such as power consumption rather than Boolean data. Among all the known side channel attacks the differential power analysis appears as one of the most efficient. This attack constitutes the main topic of this paper. More precisely, a design oriented modelling of the syndrome (signature) obtained while performing Differential Power Analysis of Kocher is introduced. As a validation of this model, it is shown how it allows identifying the leaking nets and gates during the logical synthesis step. The technology considered herein is a 130nm process. © Springer-Verlag Berlin Heidelberg 2007.