基于凸优化理论的D2D通信系统资源分配研究

发布时间：2018-04-27 21:04

本文选题：D2D通信 + 联合资源分配　；参考：《山东大学》2017年硕士论文

【摘要】：智能移动设备的普及为现代生活带来方便的同时也造成频谱资源匮乏的现状。为提高传统蜂窝网络的频谱利用率,邻近用户通过重用蜂窝频谱直接传输信息的非正交模式终端直通(device-to-device,D2D)通信技术引起研究人员的关注。如何将频谱资源块、功率等资源合理地分配给蜂窝用户与D2D用户,提高整个蜂窝D2D通信系统的传输速率与资源利用率,是资源分配策略研究的重点。另一方面,由于网络环境的恶化和移动用户数量的增多,网络安全问题变得日益重要;而通信系统自身或外部环境的限制使得信道参数的不确定性普遍存在,信道状态信息不完全可知时的鲁棒资源分配问题成为研究的热点之一。因此,针对非正交模式的蜂窝网络下的D2D通信系统,论文分别在理想信道状况无窃听和有窃听以及非理想信道状况无窃听的场景下,兼顾蜂窝子系统与D2D子系统的性能,进行问题规划和求解,并对提出的资源分配策略进行仿真分析。论文的主要工作如下:(1)简述蜂窝D2D通信系统资源分配研究的相关背景及数学理论基础,主要包含D2D通信技术、物理层安全、鲁棒性优化、凸优化理论等。(2)在信道状态信息完全可知的状况下,分别研究D2D通信系统无窃听和有窃听的联合资源分配问题。以最大化D2D子系统的非保密速率和或保密速率和为目标函数,兼顾蜂窝子系统的速率服务质量要求,联合优化蜂窝及D2D用户的功率与资源块分配,建立优化问题。进一步,把混合整数非线性的优化问题转变成凸子问题,采用拉格朗日对偶方法求得凸子问题的闭式解,并采用一维搜索方法得到原优化问题的最优解。基于以上研究,仿真分析提出的联合资源分配策略的优越性,并研究考虑D2D通信安全对系统性能带来的影响。(3)在信道状态信息不完全可知的非理想信道状况下,研究无窃听时蜂窝D2D通信系统的联合资源分配问题。采用鲁棒性优化方法,将信道状态信息中不确定的部分表示为其精确值与估计值间的有界距离,构造相应的鲁棒性优化问题。该问题中包含两类不确定参数,分别存在于目标函数与限制条件中。针对这两类不确定参数,论文把原鲁棒性问题分解为两部分,并采用最坏情况鲁棒性优化方法分别进行求解。为减少最坏情况鲁棒性优化方法产生的代价,利用机会限制方法实现鲁棒性与最优性的折衷。进一步,通过敏感度分析,研究保护函数如何对D2D子系统的和速率产生影响,并且利用数值仿真分析考虑鲁棒性时系统性能受到的影响。
[Abstract]:The popularity of intelligent mobile devices not only brings convenience to modern life, but also results in the scarcity of spectrum resources. In order to improve the spectrum efficiency of traditional cellular networks, non-orthogonal mode terminals (NCPTs), which reuse cellular spectrum to transmit information directly, have attracted the attention of researchers. How to allocate the spectrum resource blocks and power to the cellular and D2D users reasonably and improve the transmission rate and resource utilization ratio of the whole cellular D2D communication system is the focus of the research on resource allocation strategy. On the other hand, due to the deterioration of the network environment and the increase of the number of mobile users, the network security problem becomes increasingly important, and the limits of the communication system itself or the external environment make the uncertainty of channel parameters widespread. The problem of robust resource allocation when the channel state information is not fully known has become one of the hot topics. Therefore, for the D2D communication system under non-orthogonal mode cellular network, the performance of cellular subsystem and D2D subsystem are considered in the scenario of no eavesdropping and eavesdropping in ideal channel condition and no eavesdropping in non-ideal channel condition, respectively. Problem planning and solution are carried out, and the proposed resource allocation strategy is simulated and analyzed. The main work of this paper is as follows: (1) the background and mathematical theory of the research on resource allocation of cellular D2D communication system are briefly described, including D2D communication technology, physical layer security, robustness optimization, etc. Convex optimization theory et al. (2) under the condition that the channel state information is completely known, the joint resource allocation problem without eavesdropping and eavesdropping in D2D communication systems is studied respectively. The objective function is to maximize the unsecure rate and or the secure rate sum of the D2D subsystem, and to consider the rate quality of service requirements of the cellular subsystem, and to jointly optimize the power and resource block allocation of the cellular and D2D users, so as to establish the optimization problem. Furthermore, the mixed integer nonlinear optimization problem is transformed into convex subproblem, the closed solution of convex subproblem is obtained by Lagrange duality method, and the optimal solution of the original optimization problem is obtained by one-dimensional search method. Based on the above research, the superiority of the proposed joint resource allocation strategy is analyzed by simulation, and the influence of D2D communication security on system performance is studied. The joint resource allocation problem of cellular D2D communication system without eavesdropping is studied. The uncertain part of the channel state information is represented as the bounded distance between the exact value and the estimated value by using the robustness optimization method, and the corresponding robustness optimization problem is constructed. There are two kinds of uncertain parameters in the problem, which exist in the objective function and the restriction condition, respectively. For these two kinds of uncertain parameters, the original robustness problem is decomposed into two parts, and the worst-case robustness optimization method is used to solve the problem separately. In order to reduce the cost of the worst-case robust optimization method, a tradeoff between robustness and optimality is achieved by using the opportunity-constrained method. Furthermore, through sensitivity analysis, the effect of protection function on the sum rate of D2D subsystem is studied, and the system performance is analyzed by numerical simulation when robustness is considered.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【相似文献】