研究与开发
基于矩阵完备的低复杂度毫米波
大规模 MIMO 信道估计
邱佳锋
（杭州电子科技大学通信工程学院，浙江 杭州 310018）
摘 要：针对毫米波大规模多入多出系统为取得可靠信道估计需要高计算复杂度的缺陷，结合毫米波信道固
有的低秩性提出一种低复杂度的毫米波波束域信道估计算法。该算法联合利用毫米波信道波束域稀疏性和天
线域低秩性以更短训练间隔实现精确恢复。将信道估计看作矩阵完备问题，利用交替方向乘子法（ADMM）
实现快速收敛，并采用快速随机奇异值阈值法将相应计算复杂度降低一个数量级。仿真结果表明，推荐算法
在获得令人满意信道估计性能的同时依旧可以保持低复杂度。
关键词：矩阵完备；ADMM；信道估计；毫米波大规模多入多出；低复杂度
中图分类号：TN911.22
文献标识码：A
doi: 10.11959/j.issn.1000−0801.2020051
Low-complexity massive MIMO channel estimation for
mmWave systems via matrix completion
QIU Jiafeng
School of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310018, China
Abstract: Due to the high computational complexity of mmWave massive MIMO systems realizing reliable chan-
nel estimation, a low-complexity mmWave beamspace channel estimation algorithm was proposed utilizing the
inherent low-rank of the mmWave channel. The proposed algorithm jointly exploited the channel sparsity in the
beamspace domain and its low-rank property in the antenna domain to provide more accurate recovery, especially
for shorter training intervals. By regarding the channel estimation as matrix completion problem, the proposed al-
gorithm was based on the alternating direction method of multipliers with fast convergence properties and ex-
ploited fast randomized singular value thresholding to significantly reduce the corresponding complexity. The si-
mulation result shows that the proposed algorithm can achieve satisfactory channel estimation performance while
maintaining low complexity.
Key words: matrix completion, ADMM, channel estimation, mmWave massive MIMO, low complexity
收稿日期：2019−10−21；修回日期：2019−12−18
基金项目：浙江省自然科学基金资助项目（No.LY12F01008）
Foundation Item: The Natural Science Foundation of Zhejiang Province of China (No.LY12F01008)
剩余7页未读，继续阅读
资源评论
