一种自适应波束赋形的大规模MIMO信道估计方法

摘要

Numerous research results showed that massive multiple-input-multiple-output (Massive MIMO) channel exhibited a sparse structure.Using this feature,a new channel estimation algorithm had been developed,which could adaptively beam form and jointly optimize the sparse vector and matrix functions.The key part of this algorithm was to randomly optimize the structural model based on continuous constants and to use it alternately with the basic denoising optimization scheme to find the sparse feature channel.The simulation results showed that this improved sparse channel estimation method could not only reduce the number of pilots but also improve the channel estimation error of at least 20 dB compared with the common sparse channel estimation method based on Fourier transform.%大量研究结果表明大规模多输入多输出(Massive MIMO)信道表现出一种稀疏结构特性.利用这个特性开发了一种全新的信道估计算法,它能够自适应波束赋形,并且联合优化稀疏矢量和矩阵函数的方法.该算法关键部分是随机优化基于连续常量的结构模型,并且与基本的去噪优化方案交替使用以找到稀疏特征信道.仿真结果表明,与基于傅里叶变换的普通稀疏信道估计方法进行比较,这种改进型的稀疏信道估计方法不但能够允许适当减少导频数量,且至少改善20 dB以上的信道估计误差.