Research And Design Of Context-aware In Pervasive Computing Environment

In smart space, lots of contexts will be produced when dynamic user interact with applications. Conflicts will appear between contexts when users share the applications. Assigning the resources reasonably to satisfy most users in the smart space is a focal point in context computing research. The research is on how to make good use of these contexts in pervasive environment, and the contexts could help the resolution. The main research task including: detecting and collecting contexts, storage and management, and the availability of contexts.Our research is about context aware system in pervasive environment, and we simulated a smart space. Based on the context computing, we research on context modeling, and context conflicts resolution. Our primary job is as follow:We design the context model and system architecture to use in the smart space, followed with simulation. The architecture shows effectively on detecting and resolving conflicts in pervasive environment, when dynamic users access to multiple applications. And we mark off the boundary between automatic and mediated resolution, for they have different duties on solving conflicts. Automatic resolution distinguishes and filtrates the conflicts, while mediated resolution mobilizes users to make decision.In the paper, we raised an automatic resolution which can submit advanced contexts information to following requirements. The automatic system should collect all the contexts among users and applications while analyze the information and pick-up the important information. We use OWL and Jess to reason the resolution. Rules are made to turn the ambiguity into understandable information.Mediated resolution allows users to resolve conflicts in context-aware applications. However, previous works require a heavy dependence on users'intervention which may disturb users in a dynamic environment. In this paper we propose a mediated conflict resolution with a balance module. The module evaluates all the participants'feeling as a group and controls the mediation. A predicted choice is also advance to shorten the mediation time. We prove the effectiveness of our resolution through the prototype evaluation, in which our approach can resolve conflicts and reduce disturbance with appropriate parameters.In this paper, we have presented the design, prototype and evaluation study of context modeling, system architecture, and conflict resolution. The raised important technical and algorithm can be used by future investigation.