摘要
为缓解现有大规模MIMO系统集中式基带处理架构带来的带宽和计算瓶颈,DBP架构被提出。面向6G超大规模MIMO系统,提出一种DBP架构下基于EP算法的分布式信号检测算法,通过引入符号置信度构建消息传递规则并利用高斯函数近似降低消息传递算法计算复杂度,根据因子图设计中心处理单元中的消息混合规则。仿真结果表明,所提分布式EP算法性能优于现有分布式信号检测算法。
To alleviate the bandwidth and computation bottlenecks caused by the centralized baseband processing architecture in the existing massive multiple-input multiple-output(MIMO) systems, decentralized baseband processing(DBP) architecture has been recently proposed. In this paper, a decentralized signal detection algorithm based on expectation propagation(EP) under the DBP architecture is proposed for 6G ultra-massive MIMO systems. Specifically, the message passing rule is established by introducing the symbol belief, and the computational complexity of message passing algorithm is reduced via Gaussian function approximations. The message fusion rule at the centralized processing unit is designed based on the factor graph. Simulation results show that the proposed decentralized EP algorithm outperforms the state-of-the-art decentralized signal detection algorithms.
作者
张振宇
邹润民
惠峥
闫甜甜
张雨轩
戴晓明
ZHANG Zhenyu;ZOU Runmin;HUI Zheng;YAN Tiantian;ZHANG Yuxuan;DAI Xiaotning(University of Science and Technology Beying,School of Computer and Communication Engineering,Beijing 100083,China;Central South University,School of Automation,Changsha 410083,China)
出处
《移动通信》
2021年第4期16-20,共5页
Mobile Communications
基金
国家自然科学基金资助项目(61871029)。
