Felkner, A; Kozakiewicz, A
Most of the traditional access control models, like mandatory, discretionary and role based access control make authorization decisions based on the identity, or the role of the requester, who must be known to the resource owner. Thus, they may be suitable for centralized systems but not for decentralized environments, where the requester and service provider or resource owner are often unknown to each other. To overcome the shortcomings of traditional access control models, trust management models have been presented. The topic of this paper is three different semantics (set-theoretic, operational, and logic- programming) of RTT, language from the family of role-based trust management languages (RT). RT is used for representing security policies and credentials in decentralized, distributed access control systems. A credential provides information about the privileges of users and the security policies issued by one or more trusted authorities. The set-theoretic semantics maps roles to a set of sets of entity names. Members of such a set must cooperate in order to satisfy the role. In the case of logic-programming semantics, the credentials are translated into a logic program. In the operational semantics the credentials can be established using a simple set of inference rules. It turns out to be fundamental mainly in large- scale distributed systems, where users have only partial view of their execution context. The core part of this paper is the introduction of time validity constraints to show how that can make RTT language more realistic. The new language, named RT+T takes time validity constraints into account. The semantics for RT+T language will also be shown. Inference system will be introduced not just for specific moment but also for time intervals. It will evaluate maximal time validity, when it is possible to derive the credential from the set of available credentials. The soundness and completeness of the inference systems with the time validity constraints with respect to the set-theoretic semantics of RT+T will be proven.