Perturbing and protecting a traceable block cipher

10Citations
Citations of this article
21Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

At the Asiacrypt 2003 conference, Billet and Gilbert introduce a block cipher, which, to quote them, has the following paradoxical property: it is computationally easy to derive many equivalent distinct descriptions of the same instance of the block cipher; but it is computationally difficult, given one or even many of them, to recover the so-called meta-key from which they were derived, or to find any additional equivalent description, or more generally to forge any new untraceable description of the same instance of the block cipher. They exploit this property to introduce the first traceable block cipher. Their construction relies on the Isomorphism of Polynomials (IP) problem. At Eurocrypt 2006, Faugère and Perret show how to break this scheme by algebraic attack. We here strengthen the original traceable block cipher against this attack by concealing the underlying IP problems. Our modification is such that our description of the block cipher now does not give the expected results all the time and parallel executions are used to obtain the correct value. © IFIP International Federation for Information Processing 2006.

Cite

CITATION STYLE

APA

Bringer, J., Chabanne, H., & Dottax, E. (2006). Perturbing and protecting a traceable block cipher. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4237 LNCS, pp. 109–119). Springer Verlag. https://doi.org/10.1007/11909033_10

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free