Section: New Results

Verification of Security Protocols with Lists in the Symbolic Model

Participants : Bruno Blanchet, Miriam Paiola.

The symbolic model of protocols, or Dolev-Yao model is an abstract model in which messages are represented by terms. Our protocol verifier ProVerif relies on this model. This year, we have mainly worked on the verification of protocols with lists in this model.

We designed a novel automatic technique for proving secrecy and authentication properties for security protocols that manipulate lists of unbounded length, for an unbounded number of sessions. This result is achieved by extending the Horn clause approach of the automatic protocol verifier ProVerif. We extended the Horn clauses to be able to represent lists of unbounded length. We adapted the resolution algorithm to handle the new class of Horn clauses, and proved the soundness of this new algorithm. We have implemented our algorithm and successfully tested it on several protocol examples, including XML protocols coming from web services. This work has been published in [22] and our prototype is available at http://prosecco.inria.fr/personal/bblanche/publications/BlanchetPaiolaCCS13.html .

Last year, we published a conference paper that shows that, for a limited class of protocols, if a protocol is proven secure by ProVerif with lists of length one, then it is secure for lists of unbounded length. A journal version  [50] of this paper has now been accepted.