Research On Context-aware Architecture For Smart Space

The ultimate goal of pervasive computing is integrating computing device into people’s daily life seamlessly, and the user can "anytime, anywhere" get proper information and services according to their personalized needs. Context-aware is an important direction of pervasive computing research, its goal is to provide services to users according to the changing situation of users. Smart Space------a typical application of pervasive computing, is a ideal test bed for researcher to study the pervasive computing. It’s acknowledged that Smart Space is a highly dynamic, open, intelligent distributed system. In order to enable the users in the smart space to use the services provided by the Smart Space system, the system should have the capabilities to acquire and process the contextual information related to environment and user activities. Many researchers have suggested that various entities in Smart Space can be processed by computers only if the entities are represented in levels of formality. This paper mainly includes the use of the ontology description methods on information modeling, context reasoning and context-aware architecture research, etc.With the experience of Smart Space application, we analyze and categorize various types of contextual information at first. In order to describe the entities in Smart Space, we introduce an ontology method to modeling and representing contexts for the purpose of knowledge sharing and reuse. We propose an ontology-based context model, build a general ontology library consisted of concepts and relations of fundamental contexts:User, Space, Time, Event, Device, and Security.Context reasoning refers to the process that Smart Space system makes use of the contextual information about users and environment in order to acquire and extract high-level semantics. In this paper, we propose a rule-based context reasoning mechanism which includes ontology reasoning and user-defined rule reasoning. Ontology reasoning aims to solve ontology conflict detection, expression optimization, ontology composition and knowledge validation. User-defined rule reasoning aims to facilitate high-level abstraction and mapping from low-level contextual information adapted to different domains.We put forward an architecture named OCAA------Ontology-based Context-Aware Architecture, with the help of which, we apply ontology-based approach to build formal context model, and utilize rule description language to build user-defined inference rules for User Status Inquiry Service in Smart home environment.