The Research On Context-aware System Framework And Context Inconsistency Elimination Algorithm

In recent years, along with the in-depth research in key techniques such as wireless sensor networks, human-computer interaction and intelligent computing, context-aware technology got rapid development. Context-aware computing will be integrated into people’s work and life. An "ever-present and ubiquitous" context-aware computing era is coming. Context-aware computing is a human-centered calculation model, in which the computer automatically senses context and the changes of context, and then it provides the related services to users.The main research contents of this thesis are summarized as follows:(1) For complexity and dynamism of context-aware environment, this thesis studies the context-aware system framework model which can effectively shield the characteristics of context-aware environment. Nevertheless, the majority of existing context-aware system framework models are specific application-oriented, which can’t meet the demands of integrated development environment. Based on the studies of existing common framework model of context-aware system, this thesis presents a common context-aware framework model which is based on mechanisms of context quality assessment and error recovery, and applies the model in smart home. Using the mechanisms of context quality assessment and error recovery, and also the management, coordination and scheduling of context-aware environment, this model fully enhances the credibility of context-aware applications. Moreover, it provides support for interoperability between entity objects and an effective method for context information processing. Then the context-aware system becomes more adaptable, initiative and intelligent.(2) For context inconsistencies which caused by dynamic, open, distributed, heterogeneous characteristics of information sources, this thesis studies the context information inconsistency elimination algorithm. In view of the low accuracy and low self-adaptability of existing context inconsistency elimination algorithm, on the basis of studies of existing context inconsistency algorithms, this thesis presents two context inconsistency elimination algorithms, one of which based on application layer feedback and modified evidence theory, and the other one based on application layer feedback and modified reliability allocation. The simulation results show that the proposed algorithms can effectively improve context-aware system’s adaptivity and the accuracy of context inconsistency elimination.Based on the research of common framework model of context-aware system, we build a more active, intelligent and credible context-aware system framework model, which provides support for the practical application of context-aware system. Based on the research of context inconsistency elimination algorithm, we obtain two context inconsistency elimination algorithms which have more self-adaptability and higher accuracy, which provides support for effective use of context information. The above researches provide an effective method for context information processing, improve the accuracy of context information processing, and provide solid foundation for the establishment of context-aware system’s framework model and the extensive use of context-aware system.