物联网环境下轻量级服务组合研究

发布时间：2018-01-15 17:17

本文关键词：物联网环境下轻量级服务组合研究　出处：《中国石油大学(华东)》2014年硕士论文　论文类型：学位论文

【摘要】：面向服务架构(SOA,Service-Oriented Architecture)的物联网应用是指将物理世界中的各种设备的功能服务化,以一种统一和通用的接口向外界提供其功能,能够像传统互联网Web服务一样被发现、选择及组合利用。因此对物联网海量信息服务进行整合利用方面的研究对物联网应用具有重要意义。由于Web服务是实现SOA最主要的技术,为异构资源的互联与协同工作提供了有效有段,许多研究将Web服务标准直接应用与物理设备。但是物联网服务不完全等同于传统的Web服务,主要表现在:1)设备服务与物理世界直接相关而传统的Web服务来自于虚拟实体;2)设备服务所处环境是动态变化的而传统的Web服务是静态和长期存在的;3)设备服务被部署在资源(网络带宽、内存容量及处理器的计算能力等)受限的物理设备上,不可避免会产生服务请求对有限服务资源的竞争问题,更无法满足用户对服务的个性化偏好。而且传统的WS-*标准对资源受限的设备来说过于复杂且资源消耗较高。综合以上分析,传统的互联网环境下Web服务选择及组合技术没有考虑服务资源的限制,无法满足物联网服务组合的需求。为保障在资源受限和动态变化的物联网环境中服务请求者对服务质量及用户偏好的需求,本文从资源优化的角度解决服务请求对有限服务资源的竞争问题,并在此基础上探索了一种面向物联网环境下服务组合的模型,以期望能满足用户对服务质量及个性化偏好的需求。主要工作包括:1)扩展了Web服务体系结构,主要增加了对用户请求的代理与服务调度的代理。2)对物联网服务属性进行选择基的础上建立了QoS模型,在此基础上建立了基于组合赋权法的Web服务最优选择方法。3)扩展轮询队列调度算法,建立了物联网服务调度模型;4)根据扩展的服务体系结构及服务调度模型建立了本文面向物联网服务资源优化的服务组合模型;5)利用排队论的理论方法对本文模型进行了建模与分析。模拟实验结果表明,本文的方法能够在一定程度上保持用户请求的公平性,保证高优先级用户请求的低时延及时延抖动性,能够在一定程度上满足用户请求对服务质量及用户偏好的需求。
[Abstract]:SOA Service-Oriented Architecture) the application of the Internet of things refers to the ability to service the various devices in the physical world. Providing its functionality to the outside world with a unified and universal interface, it can be found as a traditional Internet Web service. Therefore, the research on the integration and utilization of mass information services in the Internet of things is of great significance to the application of the Internet of things. Because Web services is the most important technology to implement SOA. Many researches directly apply Web service standards to physical devices, but Internet of things services are not exactly equivalent to traditional Web services. The main performance is: (1) the device service is directly related to the physical world, while the traditional Web service comes from the virtual entity; 2) the environment of the equipment service is dynamic, while the traditional Web service is static and long-term. 3) device services are deployed on physical devices with limited resources (network bandwidth, memory capacity and computing power of processors, etc.), which inevitably result in competition between service requests and limited service resources. Moreover, the traditional WS-* standard is too complex for resource-constrained devices and high resource consumption. In the traditional Internet environment, Web service selection and composition technology does not consider the limitations of service resources. Can not meet the needs of the Internet of things services composition. In order to protect the resource constraints and dynamic changes in the Internet of things service requestor's demand for quality of service and user preferences. In this paper, we solve the problem of competition between service request and limited service resources from the perspective of resource optimization, and then explore a model of service composition oriented to the Internet of things environment. In order to meet the needs of users for quality of service and personalized preferences. The main work includes: 1) extended the Web services architecture. This paper mainly adds the agent of user request and service scheduling agent. 2) establishes the QoS model on the basis of selecting the properties of Internet of things services. On this basis, an extended polling queue scheduling algorithm based on combinatorial weighting method for Web service optimal selection. 3) is established, and the Internet of things service scheduling model is established. 4) based on the extended service architecture and service scheduling model, the service composition model for the optimization of Internet of things service resources is established in this paper. 5) the model of this paper is modeled and analyzed by using the theory of queuing theory. The simulation results show that the proposed method can maintain the fairness of user requests to a certain extent. The low delay delay and jitter of high priority user requests can meet the demand of QoS and user preference to a certain extent.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.44;TN929.5

【相似文献】