基于博弈的物理层安全协作技术研究

发布时间：2018-02-07 15:01

本文关键词： 物理层安全安全速率博弈收益窃听　出处：《解放军信息工程大学》2014年硕士论文　论文类型：学位论文

【摘要】：协作通信网络能有效利用无线网络的广播特性,并通过中继节点协作得到空间分集增益,成为当前的研究热点之一。然而,中继节点和协作思路的引入也给安全性带来了极大的隐患,为其安全问题研究带来了新的挑战。物理层安全技术从无线通信的最底层特点入手,为解决协作通信网络的安全问题提供了新的思路。目前,无线物理层安全协作技术研究还存在以下问题:1)缺乏有效的动态网络自私中继的选择方法,导致发送端无法实现安全速率最大化;2)缺乏抑制中继非友好行为的选择方法,导致发送端安全速率降低;3)缺乏多源多目的节点联盟机制,导致各自安全速率损耗;4)在多节点无线通信网络中,传统博弈机制会造成节点趋于非协作策略,导致网络的瓶颈安全速率降低。针对上述问题,本课题依托国家自然科学基金项目对基于博弈的物理层安全协作技术展开研究。论文首先针对自私中继和非友好中继分别提出了相应的中继选择方法:基于两次报价博弈机制的安全中继选择方法和基于博弈的非友好中继选择方法;然后,分别针对多源多目的网络和多跳无线通信网络,研究了多个节点相互协作的物理层安全方法,解决安全速率提高问题。主要研究内容包括:1、提出一种基于两次报价博弈机制的安全中继选择方法。该方法以引入虚拟货币的形式鼓励和规范不同节点间的转发关系,并提出基于虚拟货币衡量的两次报价竞争博弈机制:规定所有中继必须进行两次报价竞争,若首次被发送端选中的中继在第二次报价时未被选中,则接受惩罚被淘汰。自私中继根据转发数据所需消耗的能量等成本决定其报价,并将提供的安全速率进行标价;当两次报价的占优策略均为上报真实成本,才能获得发送端支付的报酬。发送端根据中继所提供的安全速率和报价的多少,选择使其安全速率增益最大的中继。仿真和分析表明:本方法可以在动态无线网络场景中选择到提供最大安全速率的中继,并且当发送端功率为15mW时,其安全速率相比未考虑中继自私性的传统选择方法提高40%,保证了该网络的安全传输。2、提出一种基于博弈的非友好中继选择方法。首先,建立博弈模型,明确发送端和窃听方的收益函数和收益目标:发送端的目标是通过中继发送噪声降低窃听信道容量,窃听方的目标是通过中继发送噪声降低主信道容量。之后,分别得到发送端和窃听方收益最大时从非友好中继购买的协作功率值,以及满足非友好中继为发送端服务的信道条件;最后,发送端选择满足信道条件的非友好中继并支付报酬,获得相应的协作干扰,达到最大的安全速率。仿真结果表明,通过该方法发送端可以选择到能提供最大收益的非友好中继,并且当发送端功率为15mW时,安全速率相对未考虑中继非友好性的传统方法提高近50%。3、提出一种基于协作博弈的安全联盟组网方法。首先,将协作博弈机制中的收益分摊机制引入安全联盟组网自适应形成方法,建立发送端联盟组网模型。然后,为实现联盟方式组网,基于博弈方法将联盟组网相比非联盟时增加的总安全速率作为可转移的收益函数,平均分配给组网内各个发送端;之后,发送端遍历所有可能形成的联盟组网,得到均摊收益最大的联盟组网方式;最后,发送端自适应形成该联盟组网,亟需发送信号或相同需求下窃听信道条件最坏的发送端发信号,其余所有发送端通过发送人工噪声进行协作。仿真分析验证了该方法的公平性和有效性,当发送端功率等于20mW时,高斯信道下的网络平均安全速率相比初始状态提高1.8bits/s/Hz。4、提出一种基于演化博弈机制的物理层安全协作方法。首先,根据演化博弈机制定义策略(发送人工噪声或信号)和收益(不同策略组合下的安全速率);然后,发送端根据当前网络状态以及协作收益与平均期望收益的差值,不断进行策略调整以最大化收益;最后,通过求解获得使发送端达到协作稳定策略的条件,使网络从不稳定状态向协作稳定状态演化,提高系统的瓶颈安全速率。仿真和分析结果表明,在高斯信道条件下,相比传统博弈方法,所提方法的发送端策略稳定在协作状态,网络瓶颈安全速率可提高1bits/s/Hz。
[Abstract]:Cooperative communication network can effectively utilize broadcasting characteristics of wireless networks, and get spatial diversity gain through the relay nodes, become the focus of current research. However, the introduction of relay nodes and cooperation is also to bring great security risks, the safety problem of brings new challenges to the physical layer security technology. From the bottom of the characteristics of wireless communication, provides a new way to solve the security problem of cooperative communication network. At present, there are still the following problems of wireless physical layer security cooperation technology research: 1) lack of effective dynamic network selection method of selfishness of the relay, the sending end result can not achieve the maximum safety rate; 2) lack of choice methods of inhibiting non relay friendly behavior, leading to the sending end safety rate decreased; 3) lack of multi-source and multi destination node alliance mechanism, so their safety loss rate; 4) in a multi node The wireless communication network, the traditional game mechanism will cause the node tends to non cooperative strategy, leads to the bottleneck of the network security rate is reduced. In view of the above problems, this paper relies on the National Natural Science Foundation project to study physical layer security cooperation game based on technology. This dissertation firstly for selfish relay and non friendly relay relay selection method were put forward. The corresponding selection method and relay selection method based on non friendly game safety relay two bid game mechanism based on; then, according to the multi-source multi destination network and multi hop wireless communication network, the physical layer security method of multi nodes, to solve the security problem. To improve the rate of the main research contents include: 1, put forward a selection method based on safety relay two bid game mechanism. The method by introducing a form of virtual currency to encourage and regulate the different Forwarding relationships between the nodes, and proposes the virtual currency measure two bid competition mechanism based on the provisions of all relay must be carried out two times price competition, if the relay was first selected in the sending end second times when the offer is not selected, then accept the punishment to be eliminated. The energy cost of selfish consumption according to the transmitted data the decision of the quotation, the safety rate and will provide the price; when the two quotations are reporting the real dominant strategy cost, in order to obtain the sender pay. The sender according to relay provided safety rate and price, choose to have their safety rate maximum gain relay. The simulation and analysis show that this method can be selected to provide maximum safety relay rate in dynamic wireless network scenarios, and when the transmitter power is 15mW, the safety rate compared without considering the selfishness of relay The traditional selection method to improve 40%, to ensure the safety of the.2 transmission network, this paper proposes a friendly relay selection method based on game theory. First of all, the establishment of game model, clear the sender and the eavesdropping party benefits function and profit goals: the goal is to reduce the sending end of the Wiretap Channel Capacity through the relay transmission noise and eavesdropping party the main goal is to reduce the channel capacity through the relay transmission noise. Then, respectively, the sender and the maximum profit from eavesdropping party non cooperative power friendly relay buy value, and the non friendly relay channel conditions for sending end service; finally, the sender can meet the channel conditions of non friendly relay and pay remuneration, obtain cooperation the corresponding interference to achieve safety, maximum rate. Simulation results show that by this method the sender can choose to provide the maximum benefits of the non friendly relay, and when the sending power The rate is 15mW, the relative safety rate without considering the non traditional methods of relay friendly increase nearly 50%.3, this paper puts forward a Security Alliance Cooperative networking method based on game theory. First of all, the revenue sharing mechanism of cooperative game mechanism in the introduction of security alliance network adaptive method, establish the sender Alliance Network model. Then, for the realization of alliance network mode based on game theory will total security alliance network rate increased as compared to non League revenue function can be transferred, assigned to each transmission network within the average end; then, the sender may traverse all formed alliance network, get on the maximum profit alliance network; finally, the formation of the alliance network adaptive transmitter the worst conditions, sending Wiretap Channel to send signals to the same end or demand signal, all the rest of the sending end collaboration by sending artificial noise. Simulation and analysis verify the fairness and effectiveness of the method, when the sending power is equal to 20mW, the average rate of network security Gauss channel compared to the initial state of 1.8bits/s/Hz.4, propose a physical layer security cooperation mechanism based on evolutionary game method. First, according to the evolutionary mechanism defined strategy (artificial noise or signal transmission) and benefits (safety rate under different strategies; then, the sender) according to the difference between the current state of the network and collaboration revenue and average expected profits, constantly adjust their strategies to maximize revenue; finally, to obtain the sending end to reach the cooperation strategy by solving the stability conditions, make the network unstable state evolution to the cooperative stability state, improve system security bottleneck rate. And the simulation results show that the Gauss channel conditions, compared with the traditional game method, the sending end of the proposed method The policy is stable in collaboration, and the network bottleneck security rate can be improved by 1bits/s/Hz.

【学位授予单位】：解放军信息工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN918

【相似文献】