MIMO安全通信系统中收发机算法设计

发布时间：2018-03-26 07:19

  本文选题：MIMO系统　切入点：波束成形　出处：《浙江大学》2017年硕士论文

【摘要】：随着无线通信的发展,人们对于通信的安全性提出了越来越高的要求。物理层安全技术作为传统密码加密技术的补充与升级,成为当下的研究热点。目前,学者们主要考虑将波束成形、中继和人工噪声(ArtificialNoise,AN)这三种技术应用在物理层安全的研究中。本文主要探索这三种物理层技术在具体系统下的联合应用,针对不同的系统提出了有效的设计方案,以保障通信的安全性。本文首先介绍了物理层安全领域的研究背景和发展现状,研究了信息安全理论原理、经典窃听信道模型和凸优化的基本知识,作为后续设计的理论基础。本文还总结了目前波束成形、中继和AN这三种物理层技术在安全通信方面的应用情况,提出了改进意见和进一步探索的方向。在此基础上,本文针对两种不同的多输入多输出(Multiple Input MultipleOutput,MIMO)系统,分别实现了波束成形技术和中继、AN技术的联合应用,提出了具体的设计方案。对于单用户多窃听者的MIMO协作中继系统,本文以最大化合法接收者的信噪比(Signal to Noise Ratio,SNR)为目标,在满足能量约束和一系列安全约束的条件下联合设计波束成形向量和中继放大转发矩阵。针对合法信道和窃听信道都是理想信道状态信息(Channel State Information,CSI)的模型,提出了 基于迭代优化(Alternating Optimization,AO)和转换中继(SwitchedRelaying,SR)的两种算法。本文将这两类算法推广到了合法信道理想CSI而窃听信道非理想CSI的情况,结合S-程序法完成了具有鲁棒性的算法设计。此外,本文还探索了简化的SR算法,给出了SR方案下置换矩阵码本的设计方案,并对窃听者模型进行了简单的扩展。将所提出的算法与现有的基于半定松弛(Semi-Definite Relaxation,SDR)的AO算法、零空间法、单独优化波束成形向量的算法以及单独优化中继放大转发矩阵的算法进行对比,发现无论在窃听信道理想CSI条件下还是非理想CSI条件下,本文提出的算法相对于现有的算法和单独优化一个变量的算法而言,都具有更好的SNR性能和更低的复杂度。对于多用户多窃听者的MIMO安全通信系统,本文考虑了两种不同的假设:(1)合法信道理想CSI、窃听信道非理想CSI的情况;(2)合法信道和窃听信道均为非理想CSI的情况。本文以最大化合法用户总速率为目标,联合应用波束成形技术和AN技术,在满足能量约束和窃听概率约束的条件下,提出了具有鲁棒性的波束成形向量和AN联合设计方案。本文首先将合法用户总速率最大化问题等效为加权均方误差最小化问题,同时利用Bernstein型不等式将窃听概率约束转化为一系列线性矩阵不等式(LinearMatrix Inequality,LMI)和二阶锥(SecondOrderCone,SOC)约束,提出 了基于AO的算法。本文对两种假设下所提出的算法进行了复杂度分析和性能仿真。结果表明我们所提出的算法复杂度较低并且能够有效地抵御信道CSI估计误差对合法接收者总速率的影响,具有鲁棒性。
[Abstract]:With the development of wireless communication, people put forward higher and higher requirements for the security of communication. As a supplement and upgrade of traditional cryptographic encryption technology, physical layer security technology has become the current research hotspot. The researchers mainly consider the application of beamforming, relay and artificial noise to the research of physical layer security. In order to ensure the security of communication, effective design schemes are proposed for different systems. Firstly, this paper introduces the research background and development status of physical layer security, and studies the theory of information security. The classical eavesdropping channel model and the basic knowledge of convex optimization are used as the theoretical basis for the subsequent design. The applications of the three physical layer technologies, beamforming, relay and an, in secure communication are also summarized in this paper. On the basis of this, this paper realizes the joint application of beamforming technology and relay an technology for two different multiple Input multiple output MOMOs. For the MIMO cooperative relay system with single user and multiple eavesdroppers, the aim of this paper is to maximize the signal-to-noise ratio (SNR) of the legitimate receiver. The beamforming vector and relay amplifying and forwarding matrix are designed jointly under the condition that energy constraints and a series of security constraints are satisfied. According to the model of both legitimate channel and eavesdropping channel which are ideal channel state information channel State information (CSI), the beamforming vector and relay amplification and forwarding matrix are designed. Two algorithms based on iterative optimization optimization (AOA) and switched relaying (SRS) are proposed. In this paper, the two algorithms are extended to the case of the legitimate channel ideal CSI and the eavesdropping channel non-ideal CSI. In addition, the simplified SR algorithm is explored, and the design scheme of permutation matrix codebook under SR scheme is given. And the eavesdropper model is simply extended. The proposed algorithm is combined with the existing AO algorithm based on semidefinite relaxation Semi-Definite Relationation.Zero-space method. By comparing the algorithms of single optimized beamforming vector and single optimization of relay amplification and forwarding matrix, it is found that, whether under the ideal CSI condition or the non-ideal CSI condition of eavesdropping channel, The algorithm proposed in this paper has better SNR performance and lower complexity than the existing algorithms and single optimization algorithms. For the MIMO secure communication system with multi-user and multi-eavesdropper, the proposed algorithm has better performance and lower complexity. In this paper, we consider two different assumptions: the ideal CSI in the legal channel, the non-ideal CSI in the eavesdropping channel, and the non-ideal CSI in both the lawful channel and the eavesdropping channel. In combination with beamforming and an technology, under the condition of energy constraint and eavesdropping probability constraint, A robust beamforming vector and an joint design scheme is proposed. Firstly, the problem of maximizing the total rate of legitimate users is equivalent to the weighted mean square error minimization problem. At the same time, the eavesdropping probability constraints are transformed into a series of linear matrix inequalities (LMIs) and second order order cones (SOC) constraints by using Bernstein type inequality. The algorithm based on AO is proposed. The complexity analysis and performance simulation of the two hypotheses are carried out. The results show that the proposed algorithm has lower complexity and can effectively resist channel CSI estimation. The effect of errors on the total rate of the legitimate receiver, It has robustness.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.3
，

本文编号：1666895