Privacy Engine for Context-Aware Enterprise Application Services

Satisfying the varied privacy preferences of individuals, while exposing context data to authorized applications and individuals, remains a major challenge for context-aware computing. This paper describes our experiences in building a middleware component, the Context Privacy Engine (CPE), that enforces a role-based, context-dependent privacy model for enterprise domains. While fundamentally an ACL-based access control scheme, CPE extends the traditional ACL mechanism with usage control and context constraints. This paper focuses on discussing issues related to managing and evaluating contextdependent privacy policies. Extensive experimental studies with a production-grade implementation and real-life context sources demonstrate that the CPE can support a large number of concurrent requests. The experiments also show valuable insight on how context-retrieval can affect the privacy evaluation process.

• Type: Conference
• Year: 2008
• Where: EUC
• Authors: Marion Blount, John Davis, Maria Ebling, William F. Jerome, Barry Leiba, Xuan Liu, Archan Misra
• Keywords: Applied Computing, Context Privacy Engine, EUC 2008, Privacy, Varied Privacy Preferences