具有相干信道的5G物联网系统在反馈受限时的用户选择  

User Selection for 5G IoT System with Correlated Channel under the Condition of Limited Feedback

作　　者：王强 朱晨鸣 潘甦[2] 李子博 WANG Qiang;ZHU Chenming;PAN Su;LI Zibo(Zhongtong Service Consulting Design&Research Institute Limited Company,Nanjing 210019,China;College of Communication and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]中通服咨询设计研究院有限公司,南京210019 [2]南京邮电大学通信与信息工程学院,南京210003

出　　处：《北京邮电大学学报》2024年第5期51-58,共8页Journal of Beijing University of Posts and Telecommunications

基　　金：国家自然科学基金项目(62071244)。

摘　　要：为了解决现有用户调度算法在多输入多输出(MIMO)信道相干环境下无法达到用户容量上限的问题,提出了一种新的用户选择方法。现有算法中大多是基于MIMO信道不相干的假设,但在实际通信中信道之间可能存在相干性,且在多用户MIMO系统中,用户端仅能向基站反馈部分信道的状态信息,未充分考虑干扰残留的问题。对此,首先分析了第5代移动通信系统(5G)物联网信道相干性对用户容量上限和速率的影响,揭示了该环境下用户容量的弹性特征;随后推导出了有限反馈条件下低复杂度用户速率的表达式,并设计了以最大化用户可达速率为目标的码字选择准则。针对信道相干条件下的多用户MIMO有限反馈系统,提出了一种基于强化学习的用户选择方法,避免在每个周期内重复计算速率,且计算次数仅与调度用户的数量相关。实验结果表明,在信道相干环境下,所提算法能够调度更多用户,有效提升系统的吞吐量。In order to solve the problem of existing user scheduling algorithms being unable to reach the upper bound of user capacity in a coherent environment of multiple input multiple output(MIMO)channels,a new user selection method is proposed.Existing algorithms are mostly based on the assumption that MIMO channels are incoherent,but coherent channel could be existed in practice.Meanwhile,in the multi-user MIMO system,the user can only feedback part of channel state information to the base station,which results in insufficient consideration of the residual inter-user interference.Firstly,the influence of the Internet of things(IoT)channel coherence in the fifth generation of mobile communications system(5G)on the upper limit of user capacity and transmission rates are analyzed,and the user capacity elasticity in this case is indicated.Then,the low-complexity transmission rate under limited feedback is deduced,and the codeword selection criterion based on maximizing the user's achievable data rate is designed.For multi-user MIMO limited feedback systems with channel coherence,a user selection method based on reinforcement learning is proposed.The proposed selection method can avoid recalculating the achievable rate in each cycle,and the times for calculating the rate are only related to the number of scheduled users.The experimental results show that when the system is in the coherent channel environment,the proposed algorithm can schedule more users,so as to improve the system throughput.

关 键 词：多用户多输入多输出 有限反馈 信道相干 强化学习 用户调度 

分 类 号：TN929.53[电子电信—通信与信息系统]