An Efficient Attack on All Concrete KKS Proposals

Ayoub Otmani 1, 2 Jean-Pierre Tillich 2
1 Equipe AMACC - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : Kabastianskii, Krouk and Smeets proposed in 1997 a digital signature scheme based on a couple of random error-correcting codes. A variation of this scheme was proposed recently and was proved to be EUF-1CMA secure in the random oracle model. In this paper we investigate the security of these schemes and suggest a simple attack based on (essentially) Stern's algorithm for finding low weight codewords. It efficiently recovers the private key of all schemes of this type existing in the literature. This is basically due to the fact that we can define a code from the available public data with unusual properties: it has many codewords whose support is concentrated in a rather small subset. In such a case, Stern's algorithm performs much better and we provide a theoretical analysis substantiating this claim. Our analysis actually shows that the insecurity of the proposed parameters is related to the fact that the rates of the couple of random codes used in the scheme were chosen to be too close. This does not compromise the security of the whole KKS scheme. It just points out that the region of weak parameters is really much larger than previously thought.
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00913500
Contributor : Ayoub Otmani <>
Submitted on : Tuesday, December 3, 2013 - 7:43:58 PM
Last modification on : Thursday, February 7, 2019 - 5:28:04 PM
Long-term archiving on: Monday, March 3, 2014 - 11:40:56 PM

File

article_non_anonyme.pdf
Files produced by the author(s)

Identifiers

Citation

Ayoub Otmani, Jean-Pierre Tillich. An Efficient Attack on All Concrete KKS Proposals. Post-Quantum Cryptography - PQCrypto 2011, Nov 2011, Taipei, Taiwan. pp.98-116, ⟨10.1007/978-3-642-25405-5_7⟩. ⟨hal-00913500⟩

Share

Metrics

Record views

396

Files downloads

276