Research Of The Context Information Processing Based On Granular Computing

With the development of technology about information/Internet/networking, the pervasive computing model is becoming more and more important. As a response, context awareness technology has become a research hotspot in domestic and foreign. So far, Most developed context-aware software make assumptions that the quality of the available context information is complete and accurate. But in fact context is often incomplete and fuzzy because of the frequent change in the environment and other causes, which can lead to inappropriate actions on the part of the user. So to find some methods dealing with incomplete and fuzzy context has certain practical significance.Granular computing is a discipline which based on the research of multi-level structure way of thinking.the method of problem solving and the mode fo the information processing. Ten years, granular computing gained tremendous development. So far, granular computing has been used in many fields dealing with incomplete and fuzzy information by many researchers. The thesis tries to introduce granular computing method applied to process incomplete and fuzzy context information. Major works as follows:(1) Introduced the current status of context awareness technology and granular computing theory. In the field of context-aware, some context models, uncertain reasoning, some context system frames has been briefly introduced in this thesis. In the field of granular computing, some normal models such as rough sets, fuzzy sets and quotient space has been briefly introduced. And then a briefly introduction about the applications of granular computing in data mining and image processing was given.(2) Analyzed the characteristics of the context information in context awareness system, and then a context model based on tolerance relation is given to modeling the incomplete and inaccurate context information.(3) Presented an uncertain reasoning method based on the rough logic granular. Some restraints and limits the form of the rough logic formula was given so that the method can reasoning for the incomplete and inaccurate context information. And then a example is given to show the feasibility of this method. (4) Simulated a simple intelligent laboratory space with context awareness system used the uncertain reasoning method based on the rough logic granular. Through the simulation, this thesis show that this method has a certain ability processing incomplete and inaccurate context information. Additionally, this method show a good expansibility because it's easy to add or delete rough logic formula through the simulation.