通信学报 ›› 2022, Vol. 43 ›› Issue (4): 216-226.doi: 10.11959/j.issn.1000-436x.2022058

• 学术通信 • 上一篇    下一篇

分布式IRS辅助毫米波MU-MISO系统联合波束成形设计

李中捷, 熊吉源, 高伟, 韦金迎   

  1. 中南民族大学电子信息工程学院,湖北 武汉 430074
  • 修回日期:2022-02-22 出版日期:2022-04-25 发布日期:2022-04-01
  • 作者简介:李中捷(1974- ),男,湖北武汉人,博士,中南民族大学教授,主要研究方向为无线网络的建模与分析、网络优化、通信理论等
    熊吉源(1997- ),男,河南信阳人,中南民族大学硕士生,主要研究方向为无线通信、IRS波束成形、波束追踪等
    高伟(1996- ),男,江西景德镇人,中南民族大学硕士生,主要研究方向为无线通信等
    韦金迎(1996- ),女,壮族,广西南宁人,中南民族大学硕士生,主要研究方向为无线通信等
  • 基金资助:
    国家自然科学基金资助项目(61379028);国家自然科学基金资助项目(61671483);湖北省自然科学基金资助项目(2016CFA089);中央高校基本科研业务费专项资金资助项目(CZY19003)

Joint beamforming design for distributed IRS assisted millimeter wave MU-MISO system

Zhongjie LI, Jiyuan XIONG, Wei GAO, Jinying WEI   

  1. School of Electronic Engineering, South-Central Minzu University, Wuhan 430074, China
  • Revised:2022-02-22 Online:2022-04-25 Published:2022-04-01
  • Supported by:
    The National Natural Science Foundation of China(61379028);The National Natural Science Foundation of China(61671483);The Natural Science Foundation of Hubei Province(2016CFA089);The Fundamental Research Funds for the Central Universities(CZY19003)

摘要:

为解决非视距场景下毫米波多用户多输入单输出(MU-MISO)系统下行链路的可靠性通信问题,提出一种分布式智能反射表面(IRS)辅助多用户通信的联合波束成形设计方案。考虑功率和恒模约束,以用户加权和速率最大为目标,将基站有源波束成形和多个IRS无源波束成形联合建模为非凸优化问题。利用闭式分式规划技术解耦该联合优化问题为易于求解的等价问题。根据近似线性规则和分布式连续凸近似规则,采用非凸块坐标下降算法分别交替优化有源波束成形和无源波束成形矩阵,并给出了所提算法的收敛性证明和复杂度分析。仿真结果表明,所提算法可以快速收敛,并且与2种基线算法相比,在降低复杂度的情况下能有效地提高系统传输速率。

关键词: 智能反射表面, 毫米波通信, 波束成形, 交替优化

Abstract:

A joint beamforming design scheme for distributed intelligent reflecting surface (IRS) assisted multi-user communication was proposed to solve the reliable communication problem in the downlink of millimeter-wave multi user-multiple input single output (MU-MISO) system in non line of sight scenarios.Considering the power and constant-mode constraints, the active beamforming of the base station and passive beamforming of multiple IRS were modeled as a joint non-convex optimization problem with the objective of user weighting and sum rate maximization.A closed-form fractional programming technique was used to decouple this joint optimization problem into an easily solvable equivalent problem.A non-convex block coordinate descent algorithm was used to alternately optimize the active beamforming and passive beamforming matrices according to the prox-linear rule and the distributed successive convex approximation rule.The convergence proof and complexity analysis of the proposed algorithm were also given.The simulation results demonstrate that the algorithm can converge fast, and can effectively improve the system transmission rate with reduced complexity compared with the two baseline algorithms.

Key words: intelligent reflecting surface, millimeter wave communication, beamforming, alternating optimization

中图分类号: 

No Suggested Reading articles found!