Research On Key Technologies Of Component Oriented Middleware Based On The Context-aware Service

With the rapid development of the information technology, the computing mode of the 21^st century is undergoing profound reform. It is inevitable that the distributed mobile computing will evolve to the ubiquitous computing gradually whose final goal is fusing the information space composed of computers with the physical space in which the people are working and living in. Therefore, the people can get information service transparently according to their personal needs whenever and wherever they are. However, the pervasive environment keeps changing all the time and we must adjust the systems according to these changes to satisfy the needs of the users and utilize the systems better. So the pervasive applications must be context-aware so they can adjust their actions according to the changes and the history of the contexts automatically. In a word, the research of the context-aware technology has become one of the hotspots in the pervasive computing.At the same time, the component technologies can provide better platform compatibility, better structure openness, better system reconfigurability, better scalability and better reusability. Therefore, with the development of the component technology, it has been used in the procession of developing and designing the software architecture combing with the distributed computing and mobile computing. However, the existing component-based middlewares mainly focuse on the management of the traditional distributed components and pay no attention to the context-aware component deployment and adaptations. Therefore, we pay attention to the context-aware component management in the pervasive environment to provide the context-aware management of the components .The main contents of this dissertation are composed of two parts, the first one pays attention to how to provide a context-aware service for the component based middleware to collect and manage contexts efficiently.1. Constructing the context-aware service. This paper has firstly compared and analyzed existing modeling mechanisms of contexts and put forward an ontology based hierarchal modeling approach HOCM. This approach can share the context among the users, the devices and the services to implement semantic interaction as well as reusing the field knowledge and manage the quality of the contexts. Besides these, we put forward some rule based as well as ontology based context reasoning mechanisms which proved to be effect for the reasoning of the contexts by the tests. Based on these research, this paper put forward a context-aware service model CAS by which we can manage the contexts and provide the contexts to the component based applications.2. Management of the consistency of the contexts. This paper has analyzed the introduction of the problem of the inconsistency of the contexts and put forward an inconsistency detecting algorithm based on context matching. At the same time, to resolve the context inconsistency we put forward several resolution algorithms which discard all the inconsistency contexts, the latest contexts, the contexts with the lower certainty and the contexts with lower frequency respectively. According tests prove the inconsistency of the contexts can be resolved effectively. Based on these researches we have also discussed the impact on the efficiency of the context-aware applications by the inconsistency management of the contexts.The second part of this paper pays attention to the context-aware component deployment and component adaptations.3. Context-aware component deployment. Traditional component deployment pays no attention to the changes of the environment contexts, so this paper extends existing component model based on the CAS model and puts forward context-aware component deployment architecture CACD as well as a resource restraint component deployment algorithm RRCM. The experimental results show that, the algorithm can implement context-aware component deployment in pervasive computing with lower overhead.4. Context-aware component adaptation. Traditional component adaptations pay no attention to the changes of the environment contexts, so this paper extends existing component model based on the CAS model and puts forward context-aware component adaptation model CACAM and a context-aware adaptation algorithm CACA. The experimental results show that, the algorithm can implement context-aware component adaptation in pervasive computing with lower overhead.5. Design and implementation of the system. Based on the studies of the key technologies stated above and the component based distributed computing platform StarCCM, this paper goes on with the implementation issues of context-aware middleware StarCACCM support context-aware management for the component based applications.