Mobile agents systems allow applications to explicitly relocate themselves across the network. The mobile agent paradigm provides a number of advantages in terms of service customizability and flexibility, optimized access to distributed resources, and natural support to the deployment of applications in a mobile networking environment. However, despite the advantages provided by the mobile agent paradigm, some fundamental security issues must be addressed before mobile agents receive widespread acceptance.
The overall objective of the project is to develop a general framework that provides automated support for modeling and analysis of security issues in the use of secure mobile agent systems. Preliminary work conducted at UCSB and Georgia Tech shows that appropriate analysis techniques facilitate identification of security vulnerabilities in mobile agent systems. In particular, the researchers at UCSB and Georgia Tech have successfully applied both dynamic black-box techniques and static white-box techniques to diagnose the presence of security threats in mobile agent systems.
The proposed research involves three primary efforts. First, a framework for assessing security threats for mobile agent systems will be developed. The framework supports the high-level modeling of security issues and mechanisms, by abstracting the details of specific mobile agent technologies, and encompasses static and dynamic analysis techniques. Second, a prototype system that supports the security analyses in the framework will be designed and implemented. Third, experimental subjects will be collected and experimentation will be performed. The experiments performed will allow for evaluating the effectiveness and practicality of the proposed framework on real examples defined for existing mobile agent systems.
