A peer-to-peer semantic-based service discovery method for pervasive computing environment

Abstract

The paper proposes a novel distributed service discovery method for the pervasive computing environment. The method is based on the concept of small world, policy-based advertisement and semantic-based intelligent forwarding of service request. We utilize the policy-based proactive advertisement method to establish the service community of every node, which fully consider the node capability of computation and communication. For service beyond service community, each node maintains a few distant nodes called contacts to create a small world network for increasing the semantic coverage view. Based on the hierarchical service attribute model, we integrate three-level topology character (node level, location level and service level) in contact selection mechanism. Utilizing semantic-covered network, we realize the semantic-based service discovery. Simulation result shows that our method has better search efficiency for service with different popularity than broadcastbased method. © Springer-Verlag Berlin Heidelberg 2007.