Parameterized model checking for security policy analysis

We explain how a parameterized model checking technique can be exploited to mechanize the analysis of access control policies. The main advantage of the approach is to reason regardless of the number of users of the system in which the policy is enforced. This permits to obtain more useful results from the analysis; for instance, ensuring that sensitive permissions cannot be leaked regardless of the number of users in the system. We also briefly discuss how some heuristics make the technique scalable to handle (very) large policies. This is demonstrated by a comparative experimental evaluation with state-of-the-art tools for the analysis of access control policies.