【作者】 刘刚；

【导师】 纪红；

【作者基本信息】 北京邮电大学 ， 通信与信息系统， 2015， 博士

【摘要】 随着移动互联网和物联网技术的飞速发展,移动通信网络中接入设备数量和移动数据流量不断增长,从而推动了下一代无线通信网络的研究与开发,旨在提供更高的传输速率、更好的频谱效率与能量效率,以及更低的时延。此外,最近自环干扰消除技术的进步使全双工无线通信技术成为可能。作为全双工技术的典型应用,全双工中继可以同时在同一频段上接收和发送数据,能够有效的提高网络频谱效率。因此,全双工中继技术是下一代无线通信网络中极具前景的技术之一,已受到了学术界和工业界的广泛关注。同时,无线资源管理在提升无线网络能量效率与频谱效率、保证用户服务质量(Quality of Service, QoS)等方面都扮演着重要角色。因此,本文研究了基于全双工中继的下一代蜂窝网络(5th Generation,5G)中无线资源管理技术,提出了多种新型的资源分配算法,以高效的利用网络中无线频谱,功率,基站和中继等资源,并提高网络的能量效率和资源使用效率。本文主要贡献和创新点如下：1.针对具有共享式全双工中继的蜂窝网络,提出了一种能量高效的联合功率与载波分配算法。首先,在具有共享式全双工中继的蜂窝网络中,为基站和中继提出了一种简单而实际的传输策略,以同时对抗该网络中的多址接入干扰,多用户干扰和自环干扰。第二,以最大化网络的能量效率为目标,将该网络中的联合功率和载波分配问题建模为一个非凸的混合组合优化问题。第三,为了降低计算复杂度,分两步对其求解。在第一步中,假设已知载波分配方案,将原问题转化为关于功率分配变量的凸优化问题,然后利用丁克尔巴赫方法(Dinkelbach Method)对其求解。在第二步中,基于离散随机优化提出了一种高能效的联合功率和载波分配算法。最后,通过仿真表明：在自环干扰被充分消除的情况下,本文所提联合资源分配算法能够提高网络的能量效率；与传统无线网络类似,全双工中继网络同样存在能量效率与频谱效率的折中问题。2.在基于全双工中继的虚拟化蜂窝网络中,提出了一种分布式虚拟资源分配算法。首先,将无线网络虚拟化思想引入到具有全双工中继的蜂窝网络中,并针对该网络提出了一种虚拟化资源管理框架。在该框架中,所有基础设施提供商的无线频谱、基站、全双工中继和传输功率均被视为虚拟资源,可以被动态地分配给来自不同业务提供商的用户。第二,在考虑无线网络虚拟化基本要求的条件下,以最大化基础设施提供商的总效用为目标,将该网络中虚拟资源分配问题建模为一个非凸优化问题。第三,为了高效的求解该问题,先将其转化为一个凸优化问题,然后基于方向交替乘子法(Alternating Direction Method of Multipliers, ADMM)设计了一种分布式虚拟资源分配算法。最后,通过仿真表明：在负载均衡和非负载均衡情况下,本文所提的虚拟化全双工中继网络均能提升基础设施提供商、业务提供商和用户的效用。3. 针对虚拟化全双工中继网络提出了一种基于博弈论的虚拟资源分配机制。首先,联合考虑了无线网络虚拟化和全双工中继技术,把虚拟化全双工中继网络分为四个功能实体：频谱提供商,基站提供商,中继提供商和业务提供商。第二,将该网络中的频带和功率分配问题建模为一个三级斯坦伯格博弈(Stackelberg Game)模型。在第一级博弈中,频谱提供商根据基站提供商的频谱需求,对各个频谱进行定价；在第二级博弈中,基站提供商按需向频谱提供商租用频带,并将其分配给不同的用户。在第三级博弈中,中继提供商为两跳用户执行能量高效的功率分配。第三,利用逆向推理方法对各级博弈的子博弈完美均衡进行了分析,并设计了一种迭代算法以获得该三级博弈的斯坦伯格均衡解。最后,通过理论分析和仿真验证了本文所提博弈模型的收敛性,并说明所提资源管理机制能提高各实体的效用和网络的能量效率。更多还原

【Abstract】 With the development of mobile Internet and Internet of things, there has been a continuously growth in the number of wireless devices and the amount of mobile data. This has promoted the research and development of the next generation wireless networks, to provide higher data rate, better spectrum efficiency and energy efficiency, as well as lower latency. In addition, recent advances in self-interference cancellation techniques have enabled in-band full-duplex wireless systems. As a typical application of in-band full-duplex wireless, full-duplex relaying (FDR) systems are able to transmit and receive simultaneously in the same frequency band with high spectrum efficiency. Therefore, full-duplex relaying has been one of the promising technologies in next generation wireless networks, and has received a lot of attentions from both academia and industry.Meanwhile, radio resource management plays a very important role in energy efficiency, spectrum efficiency and quality of service (QoS) provisioning in wireless networks. For those reasons, we will focus on radio resource management technologies in next generation cellular networks with full-duplex relaying, and propose several novel resource management algorithms to allocate the radio resources (e.g., wireless spectrum, transmission power, base station, relay station, etc.) with higher energy efficiency and better resource utilization. The main contributions of this dissertation are summarized as follows:1. An energy-efficient joint power and subcarrier allocation algorithm is proposed for the cellular networks with shared full-duplex relaying.Firstly, a simple but practical transmission policy is proposed to tackle the multi-access interference, multi-user interference and self-interference at the same time for the cellular networks with shared full-duplex relaying.Secondly, in order to maximize the energy efficiency of the considered network, the problem of joint power and subcarrier allocation is formulated as a mixed combinatorial and non-convex optimization problem.Thirdly, to reduce the complexity of the considered problem, we solve it in two steps. In the first step, the considered problem is transformed to a convex problem with respect to the power allocation variables, and then the Dinkelbach method is used to perform the energy efficient power allocation under the assumption of known subcarrier allocation policy. In the second step, a joint power and subcarrier allocation algorithm is presented using discrete stochastic optimization.Finally, simulation results show that the proposed joint resource allocation algorithm is able to improve the energy efficiency of FDR networks significantly if the self-interference has been sufficiently suppressed. In addition, similar to traditional wireless networks, there is also a tradeoff between energy efficiency and spectrum efficiency in FDR networks.2. A distributed virtual resource allocation algorithm is put forward for the cellular networks with wireless virtualization and full-duplex relaying.Firstly, we introduce the idea of wireless virtualization into FDR networks and propose a virtual resource management architecture for the virtualized FDR networks. In the proposed virtualized FDR networks, in addition to radio spectrum and transmission power, both base stations (BSs) and full-duplex relay stations (RSs) from different infrastructure providers (InPs) are virtualized as virtual resources, which can be dynamically shared by users from different service providers (SPs).Secondly, with the basic requirements of wireless virtualization in mind, we formulate the problem of virtual resource allocation as a non-convex optimization problem in a centralized manner, which maximizes the total utility of all InPs in virtualized FDR networks.Thirdly, to reduce the computational complexity, we transform the original virtual resource allocation problem into a convex optimization problem, and then develop a distributed resource allocation algorithm based on alternating direction method of multipliers (ADMM).Finally, simulation results show that the utility of InPs, SPs and users can be improved in the proposed virtualized FDR networks with both uniform traffic and unbalanced traffic.3. A game-based virtual resource allocation scheme is proposed for virtualized full-duplex relaying networks.Firstly, jointly considering wireless virtualization and full-duplex relaying, we devide the virtualized full-duplex relaying networks into four parts:spectrum provider, base station provider, full-duplex relay station provider, and service provider.Secondly, the problem of spectrum and power allocation in virtualized full-duplex relaying networks is modeled as a three-stage Stackelberg game. In stage one, spectrum providers offer the price of each band to the BS providers. In stage two, the BS providers purchase bands from spectrum providers and allocate the bands to different users. In stage three, the RS provider performs the energy-efficient power allocation for two-hop users.Thirdly, we analyze the subgame perfect equilibrium for each stage using backward induction method and an iterative algorithm is develop to obtain the three-stage Stackelberg equilibrium.Finally, the convergence of the proposed game is verified through theoretical analysis and computer simulations. Also, simulation results demonstrate that the proposed resource management scheme is able to improve the utilities and energy efficiency of the considered network.更多还原