Research On Uncertainties Context Reasoning Based On Membership Cloud Theory

Pervasive computing is committed to deeply embed computers in devices rounding human. Aharmonious human-computer space is emerging along with the seamless convergence ofinformation space and the real world. As a computing-soul, context-aware computing is veryimportant premise and basic for realizing pervasive computing environment. As to the complexcontext-aware computing environment, the main problem for context-aware computing lies in datasources with isomeric form, which exist conflicts and uncertainties. It is need to represent thecontexts and the definition of unified data structure or model for realizing the sharing between thedifferent devices and applications. Furthermore, an effective method should be adapted with reasonand fuse uncertainty context for acquiring more valued high context.The main works of this dissertation are induced as follows.(1) Enumerates three techniqueswith uncertainty: probability inference, D-S evidence inference and fuzzy inference. Then analyzetheir own characteristics and applicable.(2) For the uncertainty context reasoning, it proposes amethod of uncertainty context reasoning based on the membership cloud theory and introduces themembership cloud, and then deeply researches and analyzes the method.(3) A context-awaremiddleware UbiCAM prototype system for elderly medical healthcare application in pervasivecomputing environments is designed and implemented. On the hardware part, it describes thedesign and implementation of the WSN platform Ubicell and the gateway nodes. On the softwarepart, it proposes the uncertainty reasoning based on membership cloud. Through the design anddevelopment of the UbiCAM middleware prototype system, it integrates our related research resultsinto the prototype system, and demonstrates the innovation and practicability.