Research And Implementation Of The Situation Aware Of Integrated Context In The Environment Of Mobile Cloud Computing

The people's life style is changed very much by the mobile cloud computing which as a rising computing pattern. The most important characteristic of the mobile cloud computing is that there are thousands upon thousands mobile terminals such as smartphones connect to the cloud, and the terminals have many sensors which could aware the change of the context of the physics world. This pervasive capability of context aware is combined with the powerful capability of the computing of the cloud. And the whole status of the physics world might be acquainted, so the user experience is improved much such as location service, social network, contingency measure and so on. However, the current research of the domain of the traditional context aware computing always pay attention on the aggregation of the small-scale of the context such as the inside of a smart classroom. So the current research is hard to support the integration of the large-scale and different context in the environment of the combination of the cloud and the terminal. How to deal with the large-scale context and create the integrated context situation in the environment of the mobile cloud computing became the challenge which need to be solved imminently.This subject aim at these challenge, the technology of the situation aware of integrated context in mobile cloud computing is researched deeply. A context management model which run in the environment of mobile cloud computing is provided. And three key kinds of technology about the situation aware of the integrated context in the environment of the mobile cloud computing are researched.(1) How to model the context in mobile terminal. The nicer modeling notation for context could provide the fundamental support for the research of context acquisition, transmission, aggregation. The current context modeling methods are analyzed, according to the characteristics of the subject area, a context model which have the characteristics of marker model and key-value model is provided. There are advantages of intelligibility, portability and versatility in this compositive model.(2) How to collect the context in mobile terminal. There is a strong heterogeneity in mobile terminals, and the difference of configuration between terminals is also obvious. A good method of context collection should be able to fit various mobile terminals. The current methods of context collection are analyzed, and a combination context collection method of software abstract and Client/Server model is provided, manage and process large-scale context from mobile phones.(3) How to aggregate large-scale context in cloud. In traditional context-aware computing, there are two kinds of context aggregation algorithms, one is based on rules and the other is based on use cases. How to deal with the large-scale context needs powerful computing, especially the real-time transaction. According to the cloud infrastructure, a rule-based large-scale context aggregation algorithm which based on MapReduce computing model is provided, enabling the aggregation algorithm could deal with large-scale context in cloud.Based on the above work, a mobile cloud computing oriented context management framework is implemented finally. The part which run on mobile terminals is implemented on Android platform, and the part which run on cloud is implemented on Hadoop platform. An application of traffic situation aware is developed which based on the context management framework. The availability and agility of the framework are validated by this application.