A Logic-Based Framework for Web Access Control Policies

Abstract

With the widespread use of web services, there is a need for adequate security and privacy support to protect the sensitive information these services could provide. As a result, there has been a great interest in access control policy languages which accommodate large, open, distributed and heterogeneous environments like the Web. XACML has emerged as a popular access control language, but because of its rich expressiveness and informal semantics, it suffers from a) a lack of understanding of its formal properties, and b) a lack of automated, compile-time services that can detect errors in expressive, distributed and heterogeneous policies. In this dissertation, I present a logic-based framework for XACML that addresses the above issues. One component of the framework is a Datalog-based mapping for XACML v3.0 that provides a theoretical foundation for the language, namely: a concise logic-based semantics and complexity results for full XACML and various fragments. Additionally, my mapping discovers close relationships between XACML and other logic based languages such as the Flexible Authorization Framework. The second component of this framework provides a practical foundation for static analysis of expressive XACML policies. The analysis services detect semantic errors or differences between policies before they are deployed. To provide these services, I present a mapping from XACML to the Web Ontology Language (OWL), which is the standardized language for representing the semantics of information on the Web. In particular, I focus on the OWL-DL sub-language, which is a logic-based fragment of OWL. Finally, to demonstrate the practicality of using OWL-DL reasoners as policy analyzers, I have implemented an OWL-based XACML analyzer and performed extensive empirical evaluation using both real world and synthetic policy sets.