Agent driven policy management for securing open services

  • Authors:
  • Juan Jim Tan;Stefan Poslad;Leonid Titkov

  • Affiliations:
  • University of London, London;University of London, London;University of London, London

  • Venue:
  • AAMAS '03 Proceedings of the second international joint conference on Autonomous agents and multiagent systems
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Open agent services exist as heterogeneous technologies in large scale interoperable networks. The openness of such services is aggregated into multi-heterogeneous domains where it is exposed to security threats. There are also issues of various domain control privileges and security requirements specification represented over different domains. This results in the need for managing domain specific requirements in a horizontal approach where diverse domains can interact seamlessly during runtime. Hence the use of a dynamic policy model is developed to address interactions in a Multi Agent Multi Domain (MAMD) environment. This paper investigates the use of policy specifications to provide core security for access control, and communication services and domains. Besides, a fundamental attribute of the architecture, is the use of loosely coupled security and policy management mechanisms that supports the specification, representation and control of security requirements and policies. As a result, a security policy model was defined in relation to the authors' on-going work on abstract security model and mechanisms.