Skip to Main Content
Recently, the ubiquitous computing paradigm rapidly evolves and the agent technology attracts lots of researchers applying it to various applications. In the ubiquitous environment, the multiple agent system is able to provide customized service to the users by effectively utilizing the distributed system resource. In this paper we propose a dynamic prediction-based approach for both load balancing and agent pool management. It can reduce the response time for service request by avoiding unnecessary agent migration and creation of agents. The experiments reveal that the proposed scheme significantly reduces the response time compared with the existing scheme.