| 摘要: |
| 针对超大规模多输入多输出（Extremely Large-Scale Multiple-Input-Multiple-Output，XL-MIMO）系统中的近场信道估计问题，提出了一种高效的近场信道估计算法。该算法结合了粗定位和压缩采样匹配追踪技术，旨在解决传统方法在球面波传播特性下的失配问题。首先，通过特征值分解（Eigenvalue Decomposition，EVD）构造离散球面波序列字典矩阵，并对字典进行归一化处理；然后，通过接收信号，利用能量最大化方法快速确定用户设备位置；最后，采用压缩采样匹配追踪算法，利用其高效的稀疏信号恢复能力，精准获取信道的稀疏表示。仿真结果表明，与其他信道估计算法相比，所提算法在5 dB信噪比环境下归一化均方误差（Normalized Mean Square Error，NMSE）较传统方法提升2.3 dB，表现出更强的鲁棒性和更快的收敛速度。 |
| 关键词: 超大规模多输入多输出（XL-MIMO） 近场信道估计 字典矩阵归一化 匹配追踪 |
| DOI：10.20079/j.issn.1001-893x.250107004 |
|
| 基金项目:重庆市自然科学基金创新发展联合基金（中国星网）（CSTB2023NSCQ-LZX0114） |
|
| Sparse Estimation of Near-field Channels in Extremely Large-scale MIMO Systems |
| WANG Dan,WANG Jing,LIU Yuke |
| (School of Communications and Information Engineering，Chongqing University of Posts and Telecommunication，Chongqing 400065，China) |
| Abstract: |
| For the problem of near-field channel estimation in extremely large-scale multiple-input-multiple-output (XL-MIMO) systems，an efficient near-field channel estimation algorithm is proposed.This algorithm combines coarse positioning and compressed sampling matching tracking techniques to solve the mismatch problem of traditional methods under the characteristics of spherical wave propagation.Firstly，the discrete spherical wave sequence dictionary matrix is constructed by eigenvalue decomposition (EVD)，and the dictionary is normalized.Then，by receiving the signal，the energy maximization method is used to quickly determine the position of the user equipment.Finally，the compressed sampling matching pursuit algorithm is used to obtain the sparse representation of the channel accurately by using its high efficiency sparse signal recovery ability.Simulation results show that，compared with other channel estimation algorithms，the proposed algorithm improves normalized mean square error (NMSE) by 2.3 dB in 5 dB signal-to-noise ratio environment，demonstrating stronger robustness and faster convergence speed. |
| Key words: extremely large-scale MIMO（XL-MIMO） near-field channel estimation dictionary matrix normalization matching pursuit |
《电讯技术》编辑部 