协作中继通信系统中的资源分配及物理层网络编码问题研究

发布时间：2018-01-16 05:31

本文关键词：协作中继通信系统中的资源分配及物理层网络编码问题研究　出处：《东南大学》2016年博士论文　论文类型：学位论文

【摘要】：在LTE-B (Long Term Evolution Rel.12)以及未来5G (5th Generation Mobile Networks)网络中,如何扩展基站的覆盖范围对通信行业来说始终是一个研究热点。协作中继网络可以有效扩大无线通信系统的覆盖范围,提高无线链路的通信质量,被认为是5G移动通信系统的核心技术之一。随着微蜂窝甚至微微蜂窝在5G网络中的部署应用,小区范围不断缩小,中继节点发射功率小、部署灵活等特点使得其非常适合于5G网络。因此,在未来5G网络中,协作中继技术将会起到举足轻重的作用。当前协作中继技术研究主要以系统容量最大化为优化目标,而针对网络能量效率来设计网络架构和资源分配算法的相关研究尚未深入。在考虑网络整体能效最大化的同时,用户可能要牺牲自己的利益,此时必须兼顾网络性能和自私用户的公平性,而多目标优化和博弈论是解决自私用户协作行为的有效工具。本文将在保障QoS (Quality of Service)的前提下,运用多目标优化及博弈论等数学方法,深入开展协作中继网络的资源分配算法及物理层网络编码(Physical Network Coding,PNC)问题的研究。针对非对称协作中继网络的功率分配问题,本文通过引入多目标进化算法,提出了一种多源多中继网络协作功率分配策略。首先,将单目标自由搜索算法拓展到多目标优化领域,提出了多目标自由搜索算法(Multi-Objective Free Search, MOFS),并通过性能分析及数值实验验证了该算法对多目标优化问题的有效性；其次,以最大化各源节点的有效接收信噪比为目标,设计了适用于多目标自由搜索算法的效用函数和编码方案,在中继功率受限的条件下,通过MOFS算法,实现了协作中继功率资源的有效分配：最后,仿真实验结果验证了所提出策略的有效性。同时,该功率分配策略还可根据不同业务需求,通过合理选择Pareto最优解(Pareto Optimality),以得到不同的优化配置方案,提高了网络的适用性。物理层网络编码理论的提出为协作中继网络进一步提高传输容量提供了新的思路。本文针对双向协作中继信道场景,研究了基于Lattice编码(Lattice Coding)的计算转发(Compute and forward,CPF)物理层网络编码方案,以计算速率最大化为目标,提出了一种有效的双向协作中继信道计算转发编码系数向量搜索算法。针对双向协作中继信道中计算转发编码工作方式的特点,将中继节点的编码系数向量优化问题建模为带有二次约束的整数二次规划模型；针对该优化问题的特点提出了一种提升凸松弛割平面法,该搜索算法通过提升、凸松弛、生成割平面等步骤将原优化问题转化为较易求解的新松弛规划问题,通过对松弛规划问题的求解能有效获取原问题的最优解。仿真结果表明,本文所提出算法能有效获取中继节点的最优编码系数向量。在双向协作中继信道计算转发编码方案的基础上,本文继续深入研究了上行多址接入协作中继信道(Multiple Access Relay Channel, MARC)的物理层网络编码方案,从多目标优化的角度出发,提出了一种MARC计算转发编码方案的中继节点选择及系数向量搜索策略。在MOFS算法的基础上对搜索策略进行了改进,提出了一种混合整数MOFS算法,并通过标准多目标旅行商问题(Multi-Objective Travelling Salesman Problem, MOTSP)验证了该算法的有效性；针对中继节点过剩的协作中继网络场景,通过对所有可能参与协作的中继节点进行综合考虑,从系统级的角度对网络性能进行整体优化,提出了一种基于混合整数MOFS的中继节点选择及编码系数向量搜索策略,在保障所选取的中继节点系数向量矩阵满秩的前提下,以最大化最小计算速率和最大化网络和速率为优化目标,寻优结果同时实现了中继节点的选择及计算转发编码最优系数向量的搜索。仿真结果验证了所提出算法的有效性。关于协作中继网络资源分配的已有研究成果大多是针对非对称协作中继网络。本文针对对称协作中继网络的节点自私特性,提出了一种基于非协作博弈论的带宽资源分配算法。以非协作博弈论为工具,建立了一种带有竞价机制的对称协作中继网络博弈模型(Symmetric Cooperative Relay Game, SCRG),通过建立有效的激励机制,使得自私节点能通过协作资源的共享而获得公平的回报,从而积极参与协作传输；引入分布式纳什均衡(Nash Equilibrium, NE)搜索算法,通过理论推导给出了该分布式算法的收敛条件。仿真结果表明,本文所提出的博弈方法,可以合理有效地解决对称协作中继网络的协作带宽分配问题。
[Abstract]:In the LTE-B (Long Term Evolution Rel.12) and 5G (5th Generation Mobile future Networks) in the network, how to expand the coverage of the base station is always a hotspot in the communications industry. Cooperative relay network can effectively expand the coverage of the wireless communication system, to improve the communication quality of the wireless link, is considered one of the core technology of 5G the mobile communication system. With the application of micro cellular picocell deployment and 5G network in the residential area is reducing, the relay nodes transmit power is small, flexible deployment features make it very suitable for 5G network. Therefore, in the future 5G network, cooperative relay technology will play a decisive role in the cooperative relay. The main technical research to the system capacity as an optimal target, and related research for energy efficiency of the network to design the network architecture and resource allocation algorithm is No further. In considering the whole network to maximize energy efficiency at the same time, the user may have to sacrifice their own interests, fairness at this time must take into account the network performance and the user, and the multi-objective optimization and game theory is an effective tool to solve the selfish user collaboration behavior. This paper will support QoS (Quality of Service) under the premise. The application of multi-objective optimization and game theory and other mathematical methods, carry out the cooperative relay network resource allocation algorithm and physical layer network encoding (Physical Network Coding, PNC). Research on power allocation asymmetric cooperative relay network problems, through the introduction of multi-objective evolutionary algorithm, proposed a multi relay network collaboration the power allocation strategy. Firstly, the single target free search algorithm is extended to the field of multi-objective optimization, the proposed multi-objective free search algorithm (Multi-Objective Free Search, MOFS) Then, the algorithm of multi-objective optimization is verified by performance analysis and numerical experiment; secondly, to maximize the source node effective SNR as the target, the design is applied to the free search function and encoding algorithm, the relay power constrained conditions by MOFS algorithm, realize the effective allocation of cooperative relay power resource. Finally, simulation results verify the effectiveness of the proposed strategy. At the same time, the power allocation strategy can also according to different business needs, through the reasonable choice of the optimal solution of Pareto (Pareto Optimality), in order to get the optimization of different configuration, improve the applicability of the network the proposed physical layer network encoding theory for cooperative relay network to further improve the transmission capacity. This paper provides a new way for two-way cooperative relay channel scene based on Lattice (Lattice Coding) encoding calculation (Compute and forward, CPF forwarding) physical layer network encoding scheme to calculate the rate maximization as the goal, proposed a two-way cooperative relay channel effectively computes transmission coefficient vector encoding search algorithm. Based on bidirectional cooperative relay channel encoding mode calculation of forwarding, the relay node the coefficient vector encoding optimization modeling for two times with two times the integer constraint programming model; according to the characteristics of the optimization problem is proposed for improving the convex relaxation cutting plane method, the search algorithm through ascension, convex relaxation, generating cutting plane steps such as the original optimization problem is transformed into a new relaxation planning problem is easy to solve the optimal relaxation by solving programming problems can effectively obtain the optimal solution of the original problem. The simulation results show that the proposed algorithm can effectively obtain the optimal encoding of relay nodes The coefficient vector in two-way relay channel cooperation. Based on the calculation of forward encoding scheme, in-depth study of the uplink multiple access relay channel (Multiple Access Relay Channel to the MARC, the physical layer network) encoding scheme, starting from the angle of multi-objective optimization, proposed relay selection and the coefficient vector of a MARC forward calculation encoding scheme search strategy. Based on the MOFS algorithm of search strategy is improved, we propose a mixed integer MOFS algorithm, and through a standard multi-objective traveling salesman problem (Multi-Objective Travelling Salesman Problem, MOTSP) to verify the effectiveness of the algorithm; the cooperative relay network scene relay node through excess. Comprehensive consideration of all possible cooperative relay nodes, optimization of the overall performance of the network from the viewpoint of system level, is proposed based on The relay node selection and coefficient vector encoding MOFS mixed integer searching strategy is selected in the relay node security coefficient vector matrix of full rank under the premise, to maximize the minimum and maximum network computing rate and rate as the optimization objective, the optimization result and realize the relay node selection and calculation of the optimal coefficient vector encoding forwarding the search. The simulation results verify the effectiveness of the proposed algorithm. The existing research results on cooperative relay network resources allocation is mostly based on the asymmetric cooperative relay network. The node for symmetric cooperative relay networks with selfish characteristics, proposes a bandwidth allocation algorithm for non cooperative game theory based on the non cooperative game theory. As a tool to establish a cooperative bidding mechanism with symmetrical relay network game model (Symmetric Cooperative Relay Game, SCRG), through the establishment of effective The incentive mechanism, the selfish nodes can get a fair return through the collaborative sharing of resources, and actively participate in the introduction of distributed cooperative transmission; Nash equilibrium (Nash Equilibrium, NE) search algorithm is given by theoretical derivation of the convergence of a distributed algorithm. The simulation results show that the method proposed in this paper, the game, cooperative bandwidth the distribution can solve the symmetric cooperative relay network reasonably and effectively.

【学位授予单位】：东南大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN929.5

【相似文献】