超蜂窝网络架构下的高能效资源优化和传输机制  

Energy efficient resource optimization and transmission strategies for hyper-cellular mobile communication systems

作　　者：韩圣千[1] 杨晨阳[1] 李烨[2,3] 冯钢[2] 武刚[2] 郑福春[4,5] 

机构地区：[1]北京航空航天大学电子信息工程学院,中国北京100191 [2]电子科技大学通信抗干扰技术国家级重点实验室,中国成都611731 [3]School of Electrical and Computer Engineering,Georgia Institute of Technology,Atlanta GA 30332,USA [4]东南大学移动通信国家重点实验室,中国南京210096 [5]哈尔滨工业大学(深圳)电子与信息工程学院,中国深圳518055

出　　处：《中国科学：信息科学》2017年第5期560-575,共16页Scientia Sinica(Informationis)

基　　金：国家重点基础研究发展计划(973)(批准号:2012CB316003)资助项目

摘　　要：本文对我们过去5年来基于超蜂窝网络架构的高能效资源优化和传输机制的研究进展进行概述.针对第五代(5G)移动通信系统的关键技术,包括大规模集中/分布式天线、超密集网络、终端直传和协作多点等,总结系统能效–谱效之间的理论关系及其与时间、频率、天线和存储等物理资源的内在联系,并介绍面向信道与业务需求动态变化的高能效按需服务方法.研究结果表明,所有5G代表性技术的能效–谱效均呈现钟形曲线关系,即谱效较小时二者同时增长,否则两者存在折中关系;谱效优先的传输策略往往会较大程度地降低能效,但能效优先传输策略的谱效损失往往较小;多天线系统的能效随天线数渐近趋于零;超密集网络下全频复用在基站与用户密度之比较大时能效最高,否则部分频率重用能效更高;在基站部署缓存可以提升网络能效,当干扰小、回传容量低、内容流行时能效增益高.This paper provides an overview on our research on energy-efficient resource optimization and trans- mission strategies for hyper-cellular systems over the past five years. It summarizes the fundamental relationship between energy efficiency （EE） and spectral efficiency （SE）, as well as the impact of physical resources on the relationship, such as time, frequency, antenna, and cache, and reviews the energy-efficient transmission strate- gies adapted to dynamic channels and traffic. It is in the context of key technologies in fifth-generation （5G） mobile communication systems, including massive centralized/distributed antenna systems, ultra-dense networks （UDNs）, device-to-device （D2D） communications, and coordinated multi-point transmission （CoMP）. It is shown that the EE-SE relationship in all representative systems exhibits a ＂bell-shaped curve＂, meaning SE and EE increase simultaneously when SE is low while there is a tradeoff between SE and EE otherwise. ~hlrthermore, the SE loss in EE-optimal policies is low, but the EE loss in SE-optimal policies is high. EE approaches zero with an increase in the number of antennas. Universal frequency reuse can maximize EE when the base station （BS）-to-user density ratio is large while partial frequency reuse is better for EE otherwise. Caching at BSs can improve EE and EE gain is high in scenarios with weak interference, low-capacity backhaul, and large content popularity skewness.

关 键 词：能量效率 频谱效率 资源优化 超蜂窝 第五代移动通信 

分 类 号：TN929.5[电子电信—通信与信息系统]