【作者】 王羽；

【导师】 谢东亮；

【作者基本信息】 北京邮电大学 ， 计算机科学与技术， 2013， 硕士

【摘要】 随着无线传感器网络(WSN)技术内涵的扩展以及物联网和泛在网概念的出现,“信息服务”而非“连接服务”将成为未来泛在信息社会的基本特征。作为物联网的感知延伸和物联网的信息采集引擎,无线传感器网络核心价值在于可以近距离、多视角、多参数的采集环境和事件的海量、多元信息。因此,无线传感器网络是当前物联网泛在感知重要支撑网络,其稳定性高效性对信息采集效率、信息质量以及上层智能服务质量有重要作用。而如何优化泛在感知网络资源,提高网络效用,在多种优化目标中进行权衡是实现物联网面临的重要问题。网络资源中公平性问题,是经典资源分配问题。公平性最先作为经济学问题提出,九十年代后广泛应用在计算机网络性能、稳定性、平衡等优化问题中,关注多个个体竞争有限网络资源(带宽、能量、处理时间等)的资源分配问题。针对特定场景如何选取公平性概念、资源分配策略以及量化标准是其研究主要问题。本论文在深入研究无线传感器网络和公平性理论的基础上,以协同优化、跨层策略、多目标优化为技术思想,探索无线传感器网络公平性技术框架,研究公平性在WSN中解决资源分配问题；并结合特定网络环境权衡吞吐量和能量优化,以优化网络性能,提高网络效用,更好的为上层物联网智能应用服务。本文研究内容和贡献包括：1.系统性研究公平性概念定义内涵及其不同定义之间的关系；在概念基础上,分析其问题研究方法以及数学模型,分析公平性在无线网络中主要研究问题,以及在自组织网络中的应用；2.针对现有公平性研究问题限制,以协同优化、跨层策略、多目标优化为技术支撑,提出WSN公平性技术框架,为映射协议栈优化改进、多速率网络以及多目标、多层协同优化研究提供解决思路；3.结合多速率无线传感器网络场景资源问题,采用适合公平性进行网络性能优化以及结果分析,给出优化结果并量化比较。并针对无线传感网能量资源稀缺情况,在优化结果中考虑能量因素,并权衡两目标之间关系；4.根据理论分析进行协议设计和实现,在NS2仿真软件平台进行网络仿真,并对结果进行分析验证,为理论分析提供数据资料。更多还原

【Abstract】 With the expansion of wireless sensor network (WSN) technology and the emergence of the Internet of Things and ubiquitous network concept,"information service" rather than "connection service" will become the basic feature of ubiquitous information society. Wireless sensor network functions as the information collection engine and perception extension of Internet of Things. Its core value is the ability to collect massive information in a multi-angle, multi-parameter and close way. Hence, wireless sensor network is the essential supporting sensing network of Internet of Things. Its stability and high efficiency is critical for information quality and service quality of intelligent application of upper layers. Important issues facing the Internet of Things include:How to optimize the network resource, to enhance the network utility and to strike a balance between various optimization objectives.The network resources fairness problem is classic resource allocation issue. Fairness was first proposed as an economic problem, and then widely applied to the computer network performance, stability, and tradeoff optimization problems after the1990s. The key issues concern about resource allocation when multiple individuals competing for limited network resources (bandwidth, energy, processing time, etc). Main research problems in this domain include how to select specific concept of fairness, to define resource allocation strategies and to determine the quantitative criteria. On the basis of the in-depth study of the WSNs and fairness theory, combined with technical thoughts such as collaborative optimization, cross-layer strategy and multi-objective optimization, this paper explores the fairness technical framework and studies fair resource allocation problem in WSNs. The paper also takes a specific network environment into consideration, balances the optimization of energy and throughput so as to improve network performance, increase network utility and provide better service for the intelligent application services.The content and contributions in this paper include:1. Systematically study the fairness concept and the relationship between its different definitions; based on the concept, analyze research methods and mathematical models, analyze the main research problems of fairness in wireless networks, such as self-organizing network;2. Based on technical thoughts such as collaborative optimization, cross-layer strategy and multi-objective optimization, the paper propose fairness technical framework to provide possible solutions for protocol stack optimization, multirate network improvement and multi-target multi-layered collaborative optimization.3. Considering the scenario of multirate wireless sensor network, adopt proper fairness concept to optimize the network performance, analyze the optimized results and constrast with original ones in a quantitative way. Especially, in response to energy resource scarcity in WSN, the paper considers energy factor in optimized result and strikes a balance between the conflict objectives.4. Design and implement protocol in accordance with theoretical analysis. Using NS2simulation software platform for network simulation, the paper analyze simulation results to verify the theoretical analysis and provide statistical data.更多还原