面向“双碳”战略的绿色通信与网络：挑战与对策

面向国家“碳达峰”与“碳中和”的“双碳”战略需求，移动通信与网络需要在满足不断增长的业务需求前提下大幅度降低全网能耗，因此需要研究使用更少的能量传递更多信息（SMILE, send more information bits with less energy）的理论与技术。为了应对该挑战，仅靠无线传输技术的改进和硬件实现水平的提高是远远不够的，需要从系统和网络的角度探索能量的高效利用机理与方法。从能量的“节流”和“开源”2个维度展开，并针对日益增长的计算能耗给出解决方案。具体地，通过引入超蜂窝网络架构实现网络的柔性覆盖与弹性接入，使业务基站和边缘服务器在业务量较低时可以进入休眠状态，减少能量的浪费（即“节流”）。同时，大量引入可再生绿色能源（即“开源”），通过能量流与信息流的智能适配，大幅降低电网的能耗。进一步地，通过网络功能虚拟化、通信与计算资源的高能效协同，以及移动智能体的分布式计算与协同等手段，实现绿色计算与人工智能算法。

Green communication and networking for Carbon-peaking and Carbon-neutrality: challenges and solutions

Driven by the Carbon-peaking and Carbon-neutrality strategic goals, future mobile communication and networks need to drastically reduce energy consumption while satisfying the growing traffic demand. Therefore, it is necessary to study the theories and technologies used to send more information bits with less energy(SMILE). To meet the challenge, it was far from enough to rely solely on the improvements of wireless transmission technologies and hardware.It was necessary to explore efficient energy utilization mechanisms from the perspective of the system and network. Both energy saving and renewable energy utilization approaches were proposed, and solutions for the growing computing energy consumption were provided. Specifically, by introducing hyper-cellular network, agile coverage and elastic access were realized, and thus service base stations and edge servers could enter the sleeping mode to save energy when the traffic volume was low. Meanwhile, by the introduction of renewable energy and the smart adaption of energy flow and information flow, the energy consumption of the power grid could be greatly reduced. Furthermore, green computing and artificial intelligence could be realized through network function virtualization, energy-efficient coordination of communication and compute resources, and distributed computing and coordination approaches of moving agents.