可再生能源支撑云无线接入网中的资源协同调度优化策略研究

Wireless networks have experienced fast development in the past decades. Various advancing wireless technologies have been proposed to catch up with the ever-increasing diverse communication needs, one main developing trend is on the network virtualization. Analogous to computer virtualization, network virtualizaiton abstracts the physical network resources (e.g., base stations, spectrum, etc.) to allow flexible and efficient resource use and sharing. Some newly emerging representative network virtualization enabling technologies, e.g., software-defined networking (SDN), cloud-radio access networks (CRAN), have attracted much attention from both academia and industry. On the other hand, it has been widely recognized that huge energy consumption has been raised due to the massive deployment of cellular networks. Lowering the network energy consumption therefore becomes a widely concerned topic. To combat the limitations in traditional power grid, renewable energy has been widely advocated to be incorporated into radio access networks. It is promising to substitute traditional power grid by smart grid to power cellular networks. Observing the promising potential of renewable energy and CRAN in next-generation radio access networks, in this project, we are motivated to propose a holistic renewable energy powered CRAN (Re-CRAN) architecture. To further excavate its potentials, we will further conduct intensive studies from the following aspects. We will first model the achievable performance efficiency and energy efficiency by the consideration of traffic demand shape, network architecture and renewable energy generation characteristics. Based on the modeling, we will further study on the joint allocation and optimization of network resource and energy resource towards high energy-efficieny. Meanwhile, to tackle the issues of dynamically changing traffic demands and renewable generation dynamics, online scheduling algorithms will be designed. This project will make significant contribution towards next generation radio access networks. Theoretical findings and the algorithms from this project will perform as important guidance on the resource allocation and optimization in future generation wireless networks.

随着多样化通信需求以及无线接入网规模的不断增长，如何在提升网络性能与灵活性的同时，降低网络能耗，成为下一代无线网络发展的挑战。基于当前网络虚拟化技术的发展趋势，本课题提出了可再生能源支撑云无线接入网架构。针对可再生能源生产的动态随机性，在具有资源池化特性的云无线接入网中，通过网络资源与能量资源的协同调度优化，增大可再生能源利用率，是提升网络能效的关键。本课题拟从基础理论、优化模型以及算法层面对资源协同调度优化问题进行研究。首先从理论上分析可再生能源随机供给与服务质量之间的对应关系；接着围绕提升网络能效的目标，构建资源协同调度的优化模型并设计相应的资源分配算法；在此基础上，分析网络运行阶段的资源供需随机特性，设计在线资源与任务调度算法。通过本课题的研究，可望在理论上揭示资源调度与接入网服务质量之间的关联性，建立网络资源与能量资源协同调度的理论框架；从而为工程实践奠定理论基础，并提供技术支撑。

近年来，移动通信市场在用户设备（UE）连接和流量方面都出现了激增，为了满足急剧增长的通信量，下一代移动网络预计将获得比现有移动网络数千倍的容量，必然会导致较高的能源消耗和维护运营支出。为了实现下一代移动网络的节能和低成本架构，云无线接入网（C-RAN）最早由IBM提出，并由中国移动进一步开发，被广泛认为是未来接入网的关键技术。通过将一体化的宏基站拆解为负责无线信号收发的RRH和负责基带信号处理的BBU，C-RAN显著地提升了蜂窝网络的灵活性和可扩展性并降低了其运营成本。本项目研究发现解耦后的RRH能够能够由可再生绿色能源驱动，是提升接入网能效的有效途径。为此，本项目提出了可再生能源支撑云无线接入网架构，并针对可再生能源生产的动态随机性，在具有资源池化特性的云无线接入网中，通过网络资源与能量资源的协同调度优化，增大可再生能源利用率的同时，保障了用户服务质量，提升了网络能效。课题从基础理论、优化模型以及算法等层面对可再生能源支撑云无线接入网，以及相关的网络功能虚拟化，的资源协同调度优化问题进行了研究。首先从理论上基于网络微积分分析了云无线接入网的性能模型。以提升网络能效的目标，构建多资源（包括能量资源、BBU资源、RRH资源、频域资源、计算资源等）的协同调度优化模型，并设计了相应的资源分配算法。在此基础上，进一步分析了网络运行阶段的资源供需随机特性，特别是用户移动特性导致的需求侧资源变化，基于深度强化学习设计了在线的智能资源分配与任务调度算法。通过本课题的研究，在理论上揭示了资源调度与云无线接入网服务质量之间的关系，建立了网络资源与能量资源协同调度的理论框架，能够为工程实践奠定理论基础，并提供算法支撑。基于课题相关研究成果，在IEEE INFOCOM、IPDPS、SoCC、ICDCS、IWQoS等CCF A/B类国际学术会议以及IEEE TPDS、IEEE TC等领域知名国际期刊上发表论文37篇，并获得了IEEE ICPADS 2020最佳论文奖。