基于演化博弈的WSNs节点信任机制的研究与仿真

发布时间：2018-04-05 02:32

本文选题：无线传感器网络　切入点：信任管理　出处：《东华大学》2014年硕士论文

【摘要】：随着无线传感器网络的应用与发展,其安全问题受到越来越多研究者的关注。近年来,在传统安全领域之外,人们开始把目光投向WSNs的信任管理问题。 WSNs主要通过节点间互助转发的方式实现数据包传输。由于WSNs节点自身能量、资源的限制,在与其他节点配合完成任务时,往往为了减少能耗,会陷入拒绝合作的自私状态；而某些正常节点又容易被利用,帮助恶意节点转发数据包。信任管理要求建立节点间的信任机制,使得正常节点能互相信任合作,对恶意节点持不信任且拒绝合作的态度。对每个节点来说,选择信任而帮助其他节点转发数据包可以提高信誉,从而获得收益,同时会付出成本,如能量消耗、恶意信息的风险等。然而,出于节点自私性考虑,节点往往不愿帮助其他节点转发数据包从而选择不信任并拒绝合作,这会影响节点的信誉,导致收益降低。本文针对这对矛盾关系,采用演化博弈的方法,通过构建演化信任模型,引导整个WSNs达到演化稳定状态。该状态下WSNs绝大多数甚至全部节点选择信任策略。具体来说,本文的主要研究内容如下： 1.简要分析了无线传感器网络安全和信任的限制性因素,结合博弈理论对WSNs节点信任合作的矛盾点进行了阐述,并提出了本课题的研究背景、意义及主要内容； 2.针对WSNs节点非完全理性化的特点,以演化博弈为基础,构建WSNs的演化信任模型,并提出了反思与模仿两种不同的动态选择方法,通过动力学分析,推导出达到演化稳定状态的条件,并通过仿真实验验证了结论； 3.在演化信任模型中引入激励机制、丢包率和丢包重发机制,分析了激励程度、丢包情况对WSNs信任合作的影响,并通过仿真实验验证了结论。博弈理论在无线传感器网络信任管理问题上的研究受到越来越多的重视。本文基于演化博弈的方法对WSNs节点信任合作问题进行研究,揭示了WSNs节点信任演化的动力学规律,为WSNs信任管理提供了理论基础。
[Abstract]:With the development and application of wireless sensor network, its security problem attracts more and more attention of researchers. In recent years, in addition to traditional security field, people begin to pay attention to the problem of trust management of WSNs.
WSNs mainly through the way of mutual nodes forwarding packets transmission. Because WSNs nodes self energy, resource constraints, in cooperation with other nodes to complete the task, often in order to reduce energy consumption, will refuse to cooperate in selfish state; while some normal nodes are easy to be used to help malicious nodes to forward packets. Trust management requires the establishment of trust mechanism between nodes and normal nodes can make mutual trust and cooperation, distrust and refuses to cooperate on the malicious node's attitude.
For each node, the choice of trust and help other nodes to forward packets can improve the credibility, to obtain benefits, at the same time it will pay the costs, such as energy consumption, the risk of malicious information. However, for selfish considerations, nodes are often reluctant to help other nodes to forward packets to choose not to trust and refused to cooperate with the will affect the credibility of the nodes, resulting in lower earnings. In view of the contradiction between, using evolutionary game method, through building evolutionary trust model, the entire WSNs reach steady state. The state of WSNs most or all of the nodes choose trust strategy. Specifically, the main contents of this paper are as follows:
1. a brief analysis of the restrictive factors of wireless sensor network security and trust, with the contradiction of game theory to WSNs node trust and cooperation are discussed, and put forward the research background, significance and main contents;
2. according to the characteristics of non WSNs nodes completely rational, based on evolutionary game based building evolutionary trust model of WSNs, and puts forward the reflection and imitation of two different dynamic selection method, through dynamic analysis, deduced the evolution reached steady state conditions, and the conclusion is verified by simulation experiments;
3. in the evolution of trust model with incentive mechanism, packet loss rate and packet loss retransmission mechanism, analyzes the incentive degree, influence of packet loss on WSNs trust and cooperation, and verified by simulation experiment.
Study on trust management in wireless sensor network game theory has received more and more attention. This paper makes a research on the cooperation of the WSNs node trust method based on evolutionary game theory, reveals the dynamics of the WSNs node trust evolution, provides a theoretical basis for the WSNs trust management.

【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5

【参考文献】