基于演化博弈论对WSNs节点信任管理的理论研究

发布时间：2018-02-27 12:53

本文关键词： 无线传感器网络演化博弈论信任管理防御机制差额奖惩机制　出处：《昆明理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：伴随科技的不断发展与进步,无线传感器网络在生活中的各个领域都得到了大规模的应用,其中无线传感器网络的安全问题随着应用地不断扩展逐渐暴露出来。除了传统意义上的安全问题,越来越多的研究人员将重点集中在了无线传感器网络(Wireless Sensor Networks,简称WSNs)信任管理问题。由于微型传感器节点自身体积和结构的特点,导致其能量非常有限,在通信过程中,往往出于自私性的考虑会拒绝为其他节点转发数据包。并且一些正常的节点会被攻击者利用转化为恶意节点,进而恶意转发数据包,影响网络正常使用,缩短网络整体寿命。信任管理就是要在节点之间互相建立信任模型,使得正常节点之间能够正常通信,拒绝自私节点和恶意节点转发数据包的请求,使得整个网络正常使用。对于每一个节点而言,采用信任合作策略,能够增加信任值,获得信任收益,但是也会付出对应的代价,比如被恶意节点利用、自身的能量消耗等。由于节点出于自私性的考虑而拒绝为其他节点转发数据包采用不信任策略拒绝合作,这种行为将会降低信任值,导致相对应的收益降低。本文针对这种矛盾状态,通过演化博弈论构建信任管理模型,引导整个网络的大部分节点采用信任合作策略,使得整个网络最终能够正常提供服务。具体来说,本文的研究工作如下:1.本文针对存在恶意节点的无线传感器网络中进行了演化仿真,当网络中的正常节点没有采用防御策略时,该网络由于受到恶意节点的攻击,整个网络很快就瘫痪,不能继续提供正常服务。为了避免这种情况发生,本文引入了防御策略机制,在该机制下,正常节点针对恶意节点的攻击可以采用防御策略以此避免被恶意消耗,使得整个网络不至于瘫痪,即使网络在被攻击的时候,也可以提供服务。2.本文针对节点出于自私性的考虑进而成为自私节点,拒绝为其他正常节点进行数据包的转发,引入了差额奖惩机制。这种差额奖惩机制相对于传统的激励机制可以更快的使双方节点全部采用信任合作策略。在该机制作用下,可以有效降低初始采用信任策略正常节点比例数的要求。这种机制的引入使得整个网络更快地趋于稳定。3.本文针对现实网络中存在很多不可靠因素,在信任演化过程中加入了网络不可靠因素,针对发包成功率和收益延时进行了仿真分析,发现发包成功率越高越不利于网络最终的稳定,降低发包成功率和收益延时还可以降低初始采用信任策略正常节点比例数。面对以后实际的网络状况,具有一定的指导意义。本文通过演化博弈论研究了 WSNs的信任管理模型,揭示了 WSNs节点信任演化的条件规律,为后续的实际网络提供了理论基础。
[Abstract]:With the continuous development and progress of science and technology, the wireless sensor network in each field of life are widely used, the security problem of wireless sensor network with the application of expansion gradually exposed. In addition to the safety problems in the traditional sense, more and more researchers focus on the wireless sensor network (Wireless Sensor Networks, referred to as WSNs) the trust management problem. Because of the characteristics of micro sensor nodes their size and structure, due to the limited energy in the process of communication, often for selfish considerations will refuse to forward packets for other nodes. And some normal nodes will be exploited by attackers into malicious nodes and malicious data forwarding the influence of network packets, the normal use, shorten the overall network lifetime. Trust management is to establish mutual trust model between nodes So, can the normal communication between normal nodes, denial of selfish and malicious nodes forwarding the request packet, the normal use of the Internet. For each node, the trust cooperation strategy, can increase the trust value, gain trust benefits, but will also pay the corresponding price, such as by malicious nodes using their own. Energy consumption. Due to the consideration of the selfish nodes and forwarding packets by refused to distrust strategy refused to cooperate for other nodes, this behavior will reduce the trust value, resulting in the corresponding income lower. Aiming at this contradiction, through evolutionary game theory to construct the trust management model, most of the nodes of the whole network to guide the trust the cooperation strategy, the whole network can provide normal services. Specifically, the research work of this paper are as follows: 1. based on the presence of malicious nodes The wireless sensor network in the simulation, when the normal nodes in the network does not use the defense strategy, the network due to attacks by malicious nodes, the network quickly paralyzed, unable to provide normal services. In order to avoid this situation, this paper introduces the defense mechanism, in this mechanism, attacks against malicious nodes can be used as normal node defense strategy to prevent malicious consumption, making the entire network not paralyzed, even when the network is attacked, can also provide.2. services based on the consideration of the selfish nodes and become selfish nodes, to other normal nodes forwarding packets, the difference between the reward and punishment mechanism this kind of difference. The incentive mechanism with respect to the traditional incentive mechanism can be used in all trust cooperation strategy so that both nodes in the machine making faster. Use, can effectively reduce the requirements of the initial trust strategy node proportion number. The introduction of this mechanism makes the whole network quickly stabilized.3. this paper according to the reality of the network there are many uncertain factors in the evolution of trust in the process of accession to the network unreliable factor, for the success rate and return delay are analyzed the success rate is higher, that is not conducive to the stability of network, reduce the success rate and the return delay can also reduce the initial trust strategy normal nodes ratio. In the face of network situation, which has a certain guiding significance. The trust management model of WSNs is studied by evolutionary game theory, reveals the condition of WSNs the node trust evolution, provides a theoretical basis for the actual network in the future.

【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【相似文献】