通信学报 ›› 2023, Vol. 44 ›› Issue (8): 37-48.doi: 10.11959/j.issn.1000-436x.2023147

• 学术论文 • 上一篇    

Massive MIMO中通信高效的分布式预编码设计

李勉1,2,3, 李洋2,3,4, 张纵辉1,2, 史清江2,5   

  1. 1 香港中文大学(深圳)理工学院,广东 深圳 518172
    2 深圳市大数据研究院,广东 深圳 518172
    3 鹏城国家实验室,广东 深圳 518055
    4 琶洲实验室(黄埔),广东 广州 510555
    5 同济大学软件学院,上海 200092
  • 修回日期:2023-04-21 出版日期:2023-08-01 发布日期:2023-08-01
  • 作者简介:李勉(1999- ),男,江西萍乡人,香港中文大学(深圳)博士生,主要研究方向为信号处理和机器学习优化方法
    李洋(1989- ),男,吉林长春人,博士,深圳市大数据研究院副研究员,主要研究方向为无线资源管理、AI 辅助优化、大规模优化等
    张纵辉(1981- ),男,台湾桃园人,博士,香港中文大学(深圳)副教授,主要研究方向为面向无线通信、机器学习的关键信号处理和优化方法等
    史清江(1980- ),男,浙江绍兴人,博士,同济大学教授,主要研究方向为网络系统优化设计、网络/信号大数据、分布式机器学习等
  • 基金资助:
    国家重点研发计划基金资助项目(2022YFA1003900);国家自然科学基金资助项目(62071409);国家自然科学基金资助项目(62231019);国家自然科学基金资助项目(62101349);深圳市科技基金资助项目(RCJC20210609104448114);鹏城实验室重点基金资助项目(PCL2023AS1-2)

Communication-efficient distributed precoding design for Massive MIMO

Mian LI1,2,3, Yang LI2,3,4, Zonghui ZHANG1,2, Qingjiang SHI2,5   

  1. 1 School of Science and Engineering, The Chinese University of Hong Kong (Shenzhen), Shenzhen 518172, China
    2 Shenzhen Research Institute of Big Data, Shenzhen 518172, China
    3 Pengcheng Laboratory, Shenzhen 518055, China
    4 Pazhou Laboratory (Huangpu), Guangzhou 510555, China
    5 School of Software Engineering, Tongji University, Shanghai 200092, China
  • Revised:2023-04-21 Online:2023-08-01 Published:2023-08-01
  • Supported by:
    The National Key Research and Development Program of China(2022YFA1003900);The National Natural Science Foundation of China(62071409);The National Natural Science Foundation of China(62231019);The National Natural Science Foundation of China(62101349);Shenzhen Science and Technology Program(RCJC20210609104448114);Major Key Project of Pengcheng Laboratory(PCL2023AS1-2)

摘要:

针对多BBU基带处理架构,提出一种通信高效的分布式预编码方案,旨在降低BBU间前传交互和计算复杂度。首先,提出基于 R-WMMSE 算法的分布式框架,利用最优解的子空间特性无损压缩交互数据,降低数据交互量。然后设计了2种基于矩阵乘法的可学习压缩模块,通过优化的计算结构和矩阵参数减少参数和计算量,并保持函数表达能力。最后,以可达速率为优化目标,将可学习模块和分布式预编码算法框架联合优化得到最终模型。所提方案可以在更低的数据交互和计算复杂度要求下,实现预编码性能的保障。

关键词: 分布式预编码, 数据压缩, 深度学习, 联合优化

Abstract:

A communication-efficient distributed precoding scheme was proposed for multi-baseband processing unit (BBU) baseband processing architecture, aiming to reduce fronthaul data exchange and computational complexity between BBUs.Firstly, a distributed framework based on R-WMMSE algorithm was proposed, which utilized the subspace property of the optimal solution to compress the interactive data losslessly, thereby reducing data exchange.Furthermore, two learnable compression modules based on matrix multiplication were designed, using optimized computing structures and matrix parameters to reduce the parameters and computations while maintaining function expressiveness.Finally, the learnable modules and the distributed precoding framework were jointly optimized with achievable rate as the optimization objective to obtain the final model.The proposed scheme can achieve guaranteed precoding performance under lower requirements on data interaction and computational complexity

Key words: distributed precoding, data compression, deep learning, joint optimization

中图分类号: 

No Suggested Reading articles found!