多用户无线网络中的干扰对齐和物理层安全研究

发布时间：2018-06-14 15:31

本文选题：多用户无线网络 + 干扰对齐　；参考：《西安电子科技大学》2014年博士论文

【摘要】：在过去的数年间,无线网络中用户规模的快速增长及业务种类的不断丰富使得多用户干扰和无线传输安全的问题日趋严重。干扰对齐和物理层安全技术由于分别能够进行有效的干扰管理和抗窃听传输,受到了学术界和工业界的广泛关注,逐渐成为无线通信领域研究的热点。干扰对齐的核心思想是通过设计发射信号使得接收端收到的干扰尽可能地对齐在一个维数较低的子空间,同时让有用信号占据不被干扰的子空间,从而最大化自由度。物理层安全则是从信息传输的角度出发,一方面为合法用户提供可靠通信,另一方面使得窃听者无法从其截获的信号中获取任何有用消息。本文针对多用户无线网络中的干扰对齐和物理层安全分别展开研究。论文的主要工作概括如下:1.干扰对齐的研究通常考虑多输入多输出(Multiple-Input Multiple-Output,MIMO)系统,而MIMO系统在配置多根天线的同时需要多套较为昂贵的射频链路。因此,如何联合设计天线选择与干扰对齐是保证干扰管理质量的同时降低系统实现成本的关键。针对多小区MIMO蜂窝网络的下行链路,首先给出了一种利用基站控制器进行联合基站天线选择与迭代干扰对齐的处理流程,分析了面向迭代干扰对齐的最优基站天线选择方案。在此基础上,通过仿真并分析迭代干扰对齐算法的收敛特性,提出了一种基于贪心搜索策略与部分迭代机制的低复杂度基站天线选择方案。复杂度分析与仿真结果表明,该方案能够获得接近于最优基站天线选择方案的用户和速率性能,且需要的计算复杂度较低。此外,通过调整部分迭代次数,该方案可以达到用户和速率性能与计算复杂度的折中。2.现有大多数干扰对齐算法着眼于将接干扰约束并对齐到一个低维的子空间,却没有考虑有用信号的强度。针对多用户对MIMO干扰网络,为了进一步提高干扰对齐所能获得的用户和速率性能,将格拉斯曼流形优化引入到干扰对齐设计中,提出了一种新型的不需要信道互易性成立的干扰对齐算法。该算法结合极端特征值方法和格拉斯曼流形上最速下降法,同时最小化干扰子空间和干扰张成的子空间的距离,以及期望信号子空间和有用信号张成的子空间的距离,从而将干扰和有用信号分别对齐到各自相应的子空间,在对齐干扰的同时保证了有用信号的强度。仿真结果表明,该算法能够有效地增加用户和速率,其性能优于交替最小化干扰对齐以及联合有用信号的干扰对齐算法,并且通过合理地选取步长的衰减因子,可以达到用户和速率性能与收敛速度的折中,这表明该算法具有较强的灵活性。3.迭代干扰对齐需要假设本地信道状态信息(Channel State Information,CSI)和干扰协方差矩阵已知,而这些先验信息要通过复杂的训练过程才能获得。从实际工程应用的角度出发,有必要研究迭代干扰对齐的低复杂度分布式实现方法。针对多用户对MIMO干扰网络,提出了一种基于复数域快速数据投影子空间跟踪方法的低复杂度分布式迭代干扰对齐算法。在不需要本地CSI和干扰协方差矩阵先验信息的情况下,该算法直接通过干扰子空间跟踪的训练过程来迭代设计预编码矩阵和干扰抑制矩阵。复杂度分析和仿真结果表明,该算法能够以较低的计算复杂度为代价,获得接近于原始迭代干扰对齐的用户和速率性能。在此基础上,将所提出的算法进一步扩展到包含多个次级用户对的MIMO认知网络,给出了一种采用分层预编码设计的分布式迭代干扰对齐算法。该算法能够在避免次级用户对主用户产生干扰的同时,消除次级用户间的干扰。4.在无线携能通信网络中,能量接收端可能在进行无线充电的同时窃听未授权的发送给信息接收端的消息。因此,研究如何利用物理层安全技术在保证无线充电质量的前提下实现安全信息传输很有必要。针对存在多个信息接收端和能量接收端的多播场景,首先给出采用秩-2波束成形的Alamouti空时编码的传输方案,接着同时考虑信干噪比和无线充电量约束,提出了以最小化接入点发射功率为目标的基于秩-2波束成形的安全多播设计。然后,利用秩-2半定松弛技术来解决所建立的非凸优化问题,推导了秩-2半定松弛是紧松弛,也就是可以获得最优解的充分条件,并给出了得到最优解和次优解的具体求解方法。针对秩-2半定松弛不是紧松弛的情况,提出了一种能够获得次优解的秩-2高斯随机化方法。仿真结果表明,所提出的安全多播设计能够在保证信息传输安全与无线充电质量的同时,有效地降低接入点的发射功率。
[Abstract]:In the past few years, the rapid growth of user scale in wireless networks and the continuous enrichment of business types have made the problems of multiuser interference and wireless transmission security increasingly serious. Interference alignment and physical layer security technologies are capable of effective interference management and anti eavesdropping transmission, which have been widely used in academia and industry. Attention has gradually become a hot spot in the field of wireless communication. The core idea of interference alignment is to align the interference received by the receiver as far as possible to a lower dimension of the subspace by designing the transmitting signal, and to make the useful signal occupy the subspace which is not disturbed, thus maximizing the degree of freedom. On the one hand, it provides reliable communication for legitimate users, on the other hand, it makes the eavesdropper unable to obtain any useful messages from the signals captured. This paper studies the interference alignment and physical layer security in multi-user wireless networks. The main work of this paper is as follows: 1. the research on interference alignment is usually tested. Considering multiple input and multiple output (Multiple-Input Multiple-Output, MIMO) systems, the MIMO system needs multiple sets of more expensive RF links while configuring multiple antennas. Therefore, how to jointly design the antenna selection and interference alignment is the key to reduce the quality of interference management and reduce the cost of the system. A multi cell MIMO cellular network is used. In the downlink of the collaterals, first, a processing flow with base station controller for the selection of joint base station antenna and the alignment of iterative interference is given, and the optimal base station antenna selection scheme for iterative interference alignment is analyzed. On this basis, the convergence characteristics of the iterative interference to the homogeneous algorithm are simulated and analyzed, and a kind of greedy based on the greedy is proposed. The complexity analysis and simulation results show that the scheme can obtain the user and rate performance close to the optimal base station antenna selection scheme, and the computational complexity is low. In addition, the scheme can reach the user and the user by adjusting the number of iterations. Most of the existing interference alignment algorithms for rate performance and computational complexity.2. focus on the interference constraint and alignment to a low dimension subspace, but do not consider the intensity of the useful signals. For the multiuser MIMO jamming network, the Glassman flow is used to further improve the user and rate performance that the interference alignment can be obtained. The shape optimization is introduced to the interference alignment design, and a new interference alignment algorithm is proposed without the need of channel reciprocity. This algorithm combines the extreme eigenvalue method and the Glassman manifold method, and minimizes the distance between the subspace of interference subspace and the interfering tension, as well as the expected signal subspace and the useful letter. The distance of the subspace formed by the signal is aligned with the interference and useful signals to the corresponding subspace respectively, and the intensity of the useful signal is guaranteed while aligning the interference. The simulation results show that the algorithm can effectively increase the user and the rate, and its performance is better than the alternately minimized interference alignment and the interference pairs of the combined useful signal. The homogeneous algorithm, and by reasonably selecting the attenuating factor of the step length, can achieve the compromise between the user and the rate of rate performance and convergence speed, which shows that the algorithm has a strong flexibility.3. iterative interference alignment, which needs to assume the local channel state information (Channel State Information, CSI) and the interference covariance matrix known, and these prior information It is necessary to obtain a complex training process. From the perspective of practical engineering application, it is necessary to study the low complexity distributed implementation method of iterative interference alignment. A low complexity distributed iterative interference alignment algorithm based on multiple domain fast data cast shadow space tracking is proposed for multiuser MIMO interference network. Method. Without the need of local CSI and interference covariance matrix prior information, the algorithm iteratively designs the precoding matrix and interference suppression matrix through the training process of interference subspace tracking. The complexity analysis and simulation results show that the algorithm can get close to the original iteration at the cost of lower calculation complexity. On this basis, the proposed algorithm is further extended to the MIMO cognitive network containing multiple secondary users, and a distributed iterative interference alignment algorithm with hierarchical precoding is given. This algorithm can eliminate secondary users to interfere with the main user and eliminate the times. The interference.4. between the users of the level is in the wireless communication network. The energy receiver may eavesdrop on the unauthorized messages sent to the receiver at the same time. Therefore, it is necessary to study how to use the physical layer security technology to ensure the security information transmission on the premise of guaranteeing the quality of the wireless charging. The multicast scene of the information receiver and the energy receiver first gives the transmission scheme of Alamouti space-time coding using rank -2 beamforming. Then, considering the signal to noise ratio and the wireless charge quantity constraint, a secure multicast design based on rank -2 beamforming is proposed to minimize the transmission power of the access point. Then, the rank -2 half is used. In order to solve the non convex optimization problem established by the fixed relaxation technique, the sufficient condition of the rank -2 semidefinite relaxation is a compact relaxation and the optimal solution can be obtained, and a specific solution for obtaining the optimal solution and the suboptimal solution is given. A rank -2 Gauss which can obtain the suboptimal solution for the rank -2 semidefinite relaxation is proposed. The simulation results show that the proposed secure multicast design can effectively reduce the transmission power of the access point while ensuring the safety of the information transmission and the quality of the wireless charging.
【学位授予单位】：西安电子科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN92

【相似文献】