A novel approach to role-based access control

Abstract

With the rapid growth of distributed and network systems, sharing resources among many users become more common. As a result of that, we encounter with new problems concerning security and privacy on the shared resources. An access control mechanism such as role-based access control (RBAC) is one of the solutions to cope with these problems. RBAC is an efficient access control mechanism for organization data with role and permission management. In this paper, we propose a new implementation method for RBAC, which uses neural networks instead of tables. By employing neural network, it has advantages of not using multiple storages for rolepermission tables and extra mutual exclusive data tables. It also reduces access time for requested role and permission sets. © Springer-Verlag Berlin Heidelberg 2003.