Role-based access control for a distributed calculus

Rôle-based access control (RBAC) is increasingly attracting attention because it reduces the complexity and cost of security administration by interposing the notion of rôle in the assignment of permissions to users. In this paper, we present a formal framework relying on an extension of the π-calculus to study the behaviour of concurrent systems in a RBAC scenario. We define a type system ensuring that the specified policy is respected during computations, and a behavioural equivalence to equate systems. We then consider a more sophisticated feature that can be easily integrated in our framework, i.e., the possibility of automatically adding rôle activations and deactivations to processes to be run under a given policy (whenever possible). Finally, we show how the framework can be easily extended to express significant extensions of the core RBAC model, such as rôles hierarchies or constraints determining the acceptability of the system components.