大规模MIMO广播信道中分布式用户选择技术  

Distributed User Selection for Massive MU-MIMO Broadcast Channels

作　　者：薛松[1] 钱祖平[2] 钟卫[2] 邵尉[2] 

机构地区：[1]解放军理工大学通信工程学院研究生3队,江苏南京210007 [2]解放军理工大学通信工程学院

出　　处：《军事通信技术》2015年第3期61-66,共6页Journal of Military Communications Technology

基　　金：国家自然科学基金资助项目(61201241;61201218)

摘　　要：文章研究了大规模MIMO广播信道中的分布式用户选择策略,并应用迫零波束成形(ZFBF)技术与正规化迫零波束成形(RZFBF)技术抑制干扰。采用博弈论对用户进行分布式的动态选择,将其中各个用户看成是博弈的参与者。经过证明,可知博弈拥有至少一个纯策略纳什均衡(NE),最大化信道容量的最优策略组成一个纳什均衡,并提出了一个分布式算法实现纳什均衡。仿真结果表明,与已知的两个算法进行比较,通过有限反馈,可以达到最优或者渐进最优的性能,且实现简单。This paper was concerned with distributed user selection strategy for massive Multiple-Input Multiple-Output（MIMO）Broadcast Channels（BC）,where Zero-Forcing BeamForming（ZFBF）and Regularized Zero-Forcing BeamForming（RZFBF）were considered.Firstly,the selection was formulated as a discrete game in which the users are regarded as game participants.This game was shown to possess at least one pure strategy Nash Equilibrium（NE）and the optimal strategy profile which maximizes the channel capacity constitutes a pure strategy NE.Then a distributed algorithm was proposed to achieve the NE.The channel capacity performance has been improved under ZFBF and RZFBF.Compared with the two known selection algorithms,the proposed algorithm has better performance when the number of users is large in the system.Simulation results showed that the proposed algorithm,which requires limited feedback,can achieve near optimal or optimal channel capacity performance.

关 键 词：潜博弈 纳什均衡 纯策略 分布式优化 

分 类 号：TN929.52[电子电信—通信与信息系统]