On Automating the Formulation of Security Goals under the Inductive Approach uri icon

Abstract

  • The inductive approach has been successfully used for verifying a number of security protocols, uncovering hidden assumptions and even attacks. Yet it requires a high level of skill to use: a user must guide the proof process, selecting the tactic to be applied, inventing a key lemma, etc. Proofs are both onerous and cumbersome. To compound the problem, security guarantees are not built into the logic but proven within it, making it difficult to give an account of proof discovery. We introduce a method, which, given a protocol, automatically formulates most of the properties that the protocol should enjoy in order to ensure security. If proven, these guarantees can be combined to provide a story as to why the protocol achieves security. Otherwise, the unproven goals may be used to uncover hidden assumptions or attacks.

Publication date

  • December 1, 2003