普适计算中基于隶属云理论的不确定性上下文处理研究

发布时间：2018-03-07 18:34

本文选题：普适计算　切入点：上下文推理　出处：《南京邮电大学》2012年硕士论文　论文类型：学位论文

【摘要】：普适计算致力于将计算融入人们的日常生活中，将由计算和通信节点及系统组成的计算空间与人们生活的物理空间无缝地集成为和谐的人机交互信息环境。上下文感知技术是普适计算中的重要技术，，是实现普适环境的前提和基础，是普适计算的灵魂。上下文感知计算要解决的主要问题是由于环境的复杂与场景多变而使得上下文信息来源具有异构性、冲突性和不确定性特点。因此为实现不同设备和应用间的共享，需要对上下文进行统一的数据结构定义或模型描述，并采用有效的不确定性推理融合方法，以提供具有价值的高层上下文。本文的主要工作归纳如下：（1）列举现有的三种上下文推理技术：贝叶斯推理、证据推理和模糊推理，并分析各自的特点以及适用的情况。（2）针对不确定性上下文，介绍了隶属云理论，提出了基于隶属云理论的不确定性上下文推理，并进行了详细分析。（3）本文最后设计并实现了一个普适计算环境下针对老年医疗健康护理应用的情景感知中间件UbiCAM原型系统。在硬件部分，详细地介绍了作为WSN平台载体的Ubicell节点以及WSN网关节点的设计与实现；在软件部分，提出了基于隶属云理论的不确定性情景推理方法。通过该UbiCAM中间件原型系统的设计与开发，把研究成果融入其中，较好地展示了研究成果所具有的创新性和实用性。
[Abstract]:Pervasive computing is dedicated to integrating computing into people's daily lives. The computing space composed of computing and communication nodes and systems is seamlessly integrated into a harmonious human-computer interactive information environment with the physical space in which people live. Context-aware technology is an important technology in pervasive computing. It is the premise and foundation of implementing pervasive environment and the soul of pervasive computing. The main problem to be solved in context-aware computing is that the source of context information is heterogeneous due to the complexity of the environment and the changing of the scene. Therefore, in order to realize the sharing of different devices and applications, it is necessary to define the data structure or describe the model in a unified way, and adopt an effective method of uncertainty reasoning fusion. To provide a high-level context of value. The main work of this paper is summarized as follows: (1) enumerating three existing contextual reasoning techniques: Bayesian reasoning, evidential reasoning and fuzzy reasoning, and analyzing their respective characteristics and applicable conditions. The membership cloud theory is introduced, and the uncertainty context reasoning based on membership cloud theory is proposed. Finally, this paper designs and implements a situational sensing middleware UbiCAM prototype system for geriatric medical and health care applications in the context of pervasive computing. The design and implementation of Ubicell node and WSN gateway node as the carrier of WSN platform are introduced in detail. In the software part, an uncertain scenario reasoning method based on membership cloud theory is proposed. Through the design and development of the UbiCAM middleware prototype system, The innovation and practicability of the research results are well demonstrated by incorporating the research results into them.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP338;TP391.1

【参考文献】