An ontology-based P2P infrastructure to support context discovery in pervasive computing

Abstract

Pervasive computing smart spaces are rich in context information provided by various interconnected sensors and software sources distributed in multiple smart spaces. These context providers need to be efficiently and automatically tracked by the context-aware applications via the context discovery process. This dissertation addresses various issues in system infrastructure support for context discovery from multiple smart spaces. Based on the hybrid centralized-decentralized context discovery model, a context discovery platform, named Orion, is proposed to enable context provider to publish and context requester to lookup context information. We propose a Peer-to-Peer computing system approach for inter-space resource communication, Semantic Overlay Network techniques for efficient query distribution, and Semantic Web ontology modeling and reasoning techniques for context representation and semantic-based matchmaking. Orion enables discovery and retrieval of context information from multiple smart spaces, and thereby allowing flexible design of context-aware applications and dynamic use of a wide range of context information from multiple sources.