Efficient Policy Analysis for Administrative Role Based Access Control By

Role Based Access Control (RBAC) has been widely used for restricting resource access to only authorized users. Administrative Role Based Access Control (ARBAC) specifies permissions for administrators to change RBAC policies. It is often difficult to fully understand the effect of an ARBAC policy by simple inspection, because sequences of changes by different administrators may interact in unexpected ways. ARBAC policy analysis algorithms can help by answering questions, such as user-role reachability, which asks whether a given user can be assigned to given roles by given administrators. This problem is intractable in general. The object of this dissertation is to develop efficient algorithms for analysis of ARBAC policies. First, we present user-role reachability analysis algorithms for ARBAC that are fixed parameter tractable. These algorithms have a high complexity with respect to some parameter k that is often small in practice. We have measured the performance of our analysis algorithms on a university policy, a healthcare policy, and families of synthetic policies in order to validate our parameterized complexity results. In addition, we have considered other analysis problems, including role containment and weakest preconditions. Next, we present the first known incremental algorithms for analysis of evolving ARBAC policies. ARBAC policies tend to change over time in order to fix design flaws or to cope with the changing requirements of an organization. Changes to ARBAC policies may invalidate security properties that were previously satisfied. Our incremental algorithms determine if a change may affect the analysis result, and if so, reuse the information of the previous analysis to incrementally update the analysis result. Detailed evaluations show that our incremental algorithms outperform the non-incremental algorithm in terms