【作者】 林敏；

【导师】 王东；

【作者基本信息】 上海交通大学 ， 软件工程， 2011， 硕士

【摘要】 随着智能环境的发展,越来越多的传感器、移动设备、嵌入式设备出现在我们生活环境当中。无线传感网技术(WSN)、无线射频识别技术(RFID)以及各种无线连接方案日益成熟,为设备之间直接交互提供了技术支撑。应用RFID和WSN等信息采集和传输技术构建智能环境,融合了物理空间和信息空间,能为用户提供全局性服务。目前的智能环境系统主要围绕利用上下文感知和多通道信息获取技术实现普适计算应用场景,但这些系统应用流程相对固定,既难以满足用户需求各异性的需要,又缺乏对服务按需扩充的灵活性,基于服务组合技术的智能环境动态服务推送受到关注。本文结合上海市经委的《面向世博的RFID未来商店关键技术与装备研发和示范工程建设》项目,深入研究了基于RFID和WSN的智能环境服务组合技术,通过对服务建模的研究,设计了基于OWL-S的智能环境服务模型,以典型的RFID和WSN服务进行服务模型开发,应用到基于RFID和WSN的智能购物环境中,实现了一个智能环境中面向用户的动态服务组合系统。本文设计与开发了服务组合引擎,该引擎分为四个部分:任务生成模块、服务发现模块、工作流生成模块、服务执行模块。任务生成模块负责将用户自然语言请求转化为计算机可识别的目标任务;服务发现模块从服务注册器中查找、匹配符合任务请求的服务模型;工作流生成模块根据服务发现的拓扑结构生成工作流;服务执行模块按照工作流图执行服务。本文还研究了服务生命周期的管理,适应了智能环境服务动态性等特点。本文首先对智能环境发展做了综述,分析智能环境服务的特点;接着,总结分析了智能环境服务组合的需求和挑战;然后设计了面向用户的智能环境服务组合总体架构,并详细叙述各个模块的设计方案;最后,描述了智能环境服务组合系统的实现。更多还原

【Abstract】 With the development of smart environment, more and more sensors, mobile devices and embedded devices appear in our daily life. Wireless Sensor Network, Radio Frequency Identification technology and a variety of wireless connect solutions provide technical support for the direct interaction among devices. Using RFID and WSN applications as new information collection and input methods, smart environment bridge the physical and information worlds together and provide global services to users. The current smart environment systems are mainly around using context-aware technology and multi-channel information accessing technology to achieve pervasive computing application. However, these applications are relatively fixed, not only difficult to meet the users’complex needs, but also lack of flexibility to expand demand for services. At present, more and more attention is paid on service composition technology in the smart environment.In order to welcome Shanghai Expo 2010, Shanghai government proposals a project called "Future Store technology and equipment research for Shanghai Expo" into Retail Chain. In this paper, we deeply study the service composition technology in RFID and WSN based smart environment. Through the study on service modeling, we design an OWL-S based smart environment service model. Then we develop typical RFID and WSN services model based on it, and apply to the smart shopping environment. We also design and development a service composition engine, which is divided into four parts: task generation module, service discovery, workflow generation module and service execution module. Task generation module is responsible to translate natural language request into a language which computer can identify. Service discovery module discovers and matches the users’request consistent with the service model. Workflow generation module creates workflow according to service discovery topology order. Service execution module performs the services according to the workflow diagram. Besides, we also study on the service life cycle management.The contents of this paper are organized as follows: Firstly, we analyze of the characteristics of smart environmental services. And we summarize needs and challenges of the service composition technology in smart environment. Then we design a user-oriented architecture for service composition in smart environment, and describe detail of each module design. Finally, we depict the implementation of service composition system.更多还原