SGER: A Computing Paradigm for Collaborative Multiple Mobile Agent Applications

University Of Florida, Gainesville FL

Investigators

Abstract

Abstract The project is an exploratory research into the potential of using a new promising programming paradigm, which combines two agent-based concepts: mobile agent computing and multiple collaborative agent systems, for emerging distributed applications. One major obstacle that hinders the development of mobile agent systems is due to the concern of agent protection against malicious hosts. The research conjectures that the issue can be addressed and the security of agents can be enhanced by relying on the concept of protection as a whole, i.e., the secrecy and integrity of a mobile agent processes can be better protected through dispersion of its functions and data resulting in a system of multiple collaborating mobile agents. Viewing from a different angle, the research extends multiple agent systems with a new dimension in mobility. This extension will broaden the application of distributed collaborative agent systems. The goal of this research is to prove the feasibility and the holistic effect of unifying mobility and distributiveness in the design of future Internet services, and meanwhile, to identify system requirements and their supporting algorithms/protocols in this new computing paradigm. Since the proposed research is on the design of a general computing system architecture for distributed applications, its impact will be broad and potentially useful for any network-based applications that require agent collaboration. The initial system assumption is most likely to be well suited for e-commerce systems where security is an essential requirement. From the educational perspective, the project is ideal for research pursuing graduate students since it involves a broad spectrum of fundamental research topics in algorithm and theory, distributed systems, mobile and pervasive computing, and can cross into AI and language design.