SEER: Security enhanced entity-relationship model for secure relational databases

Abstract

In this paper we propose extensions of the ER modeling concepts to address the conceptual modeling of security features and authorization histories in a database. It is designed to serve as the global conceptual model in a federated database environment. In such databases, data belonging to a variety of users under different constraints and security policies is subject to sharing. For the purposes of this paper, we assume that the local databases are relational. We propose a two-layered representation of data, one of which is the traditional ER model, whereas the other layer deals with the history of authorizations. We propose a methodology for dealing with relational databases by first analyzing the local database schemas and then mapping the security constraints appropriately while transforming/integrating each local data model into the global data model. We believe that the proposed conceptual model provides a useful common framework for dealing with the modeling and enforcement of security schemes in federated database environments.