一种水平和垂直联合优化的3D波束成形算法  

A 3D-Beamforming Algorithm Based on Horizontal and Vertical Jointly Precoding Optimization

作　　者：张仕宇[1] 冯辉[1] 宗志远[1] 刘凯[1] 杨涛[1] 胡波[1,2] 

机构地区：[1]复旦大学电子工程系,上海200433 [2]复旦大学电磁波信息科学教育部重点实验室,上海200433

出　　处：《复旦学报（自然科学版）》2015年第5期569-578,共10页Journal of Fudan University：Natural Science

基　　金：国家科技重大专项(2013ZX03003006-003)

摘　　要：与传统的无源天线阵列相比,基站端采用有源天线阵列,能够实现更加灵活和智能的波束成形,充分地利用三维空间信道特性,进而提高系统性能.本文研究了单小区多用户多输入多输出(MU-MIMO)下行广播信道场景中,基于面阵有源天线阵列的基站预编码3D波束成形设计问题.首先,将3D波束成形问题建模为对水平预编码和垂直预编码的联合设计问题,并引入一种最小化基站发射功率的最优化模型.然后,将原问题分解为关于水平预编码和垂直预编码的优化子问题,并通过半定松弛和交替优化迭代方法求解.仿真结果表明本文提出的算法能够很快收敛,并且能够获得明显优于对比方案的系统和速率性能.The base station（BS）,which is equipped with the active antenna array rather than the conventional passive antenna array,can facilitate more flexible and intelligent beamforming and take full advantage of the 3-dimensional spatial channel property,resulting in better system performance.In this paper,a 3Dbeamforming scheme based on the planar active antenna array is proposed in a single-cell multiuser multiple-input multipleoutput（MU-MIMO）downlink system.First,the 3D-beamforming problem is modeled as a jointly design problem of the horizontal and vertical precoding for the BS.And a classic optimization model which minimizes the total transmitted power is introduced.Since finding the joint optimal parameters is not easy,the original optimization problem is decoupled into two sub-problems of the horizontal and vertical precoding respectively.And then the solution can be approached by semi-definite relaxation（SDR）and alternating iteration method.Simulation results show that the proposed 3D-beamforming algorithm can converge rapidly and outperforms the comparison algorithms in cell sum-rate.

关 键 词：大规模MIMO 有源天线系统 3D-MIMO 3D波束成形 

分 类 号：TN911.72[电子电信—通信与信息系统]