This paper presents applicability of Strong Stationary Times (SST) techniquesin the area of cryptography. The applicability is in three areas: *) Propositions of a new class of cryptographic algorithms (pseudo-randompermutation generators) which do not run for the predefined number of steps.Instead, these algorithms stop according to a stopping rule defined as SST, forwhich one can obtain provable properties: *** results are perfect samples from uniform distribution, *** immunity to timing attacks (no information about the resultingpermutation leaks through the information about the number of steps SSTalgorithm *) We show how one can leverage properties of SST-based algorithms toconstruct an implementation (of a symmetric encryption scheme) which is immuneto the timing-attack by reusing implementations which are not secure againsttiming-attacks. In symmetric key cryptography researchers mainly focus onconstant time (re)implementations. Our approach goes in a different directionand explores ideas of input masking. *) Analysis of idealized (mathematical)models of existing cryptographic schemes -- i.e., we improve a result byMironov ((Not So) Random Shuffles of RC4; Advances in Cryptology -- CRYPTO2002)
展开▼
