Robust Linear Precoder Design for 3D Massive MIMO Downlink With A Posteriori Channel Model
Robust Linear Precoder Design for 3D Massive MIMO Downlink With A Posteriori Channel Model
复制标题
DOI：
10.1109/tvt.2022.3163392
复制
发表时间：
2020-04
影响因子：
6.8
通讯作者：
An-an Lu;Xiqi Gao;Chengshan Xiao
中科院分区：
文献类型：
--
作者：
An-an Lu;Xiqi Gao;Chengshan Xiao
本文研究了三维（3D）大规模多输入多输出（MIMO）下行链路在均匀平面阵列（UPA）和不完全信道状态信息（CSI）下的线性预编码器设计。首先利用采样导向矢量建立了基于波束的统计信道模型（BSCM），然后建立了包含信道老化的后验信道模型。在后验信道模型的基础上，我们考虑了在总功率约束下，通过最大化期望加权和速率的上界来设计鲁棒预编码器。我们推导出目标函数的两个凹极小化函数。这些minorizing功能和minorize-maximization（MM）的方法，我们得到两个迭代算法收敛到稳定点的优化问题。仿真结果表明，在中高移动场景下，该预编码器比常规迫零（RZF）预编码器和信漏噪比（SLNR）预编码器具有更高的性能增益。
In this paper, we investigate the linear precoder design for three dimensional (3D) massive multi-input multi-output (MIMO) downlink with uniform planar array (UPA) and imperfect channel state information (CSI). We introduce a beam based statistical channel model (BSCM) by using sampled steering vectors, and then an a posteriori channel model which includes the channel aging is established. On the basis of the a posteriori channel model, we consider the robust precoder design by maximizing an upper bound of the expected weighted sum-rate under a total power constraint. We derive two concave minorizing functions of the objective function. With these minorizing functions and the minorize-maximization (MM) methodology, we derive two iterative algorithms that converge to stationary points of the optimization problem. Simulation results show that the proposed precoders can achieve a significant performance gain than the widely used regularized zero forcing (RZF) precoder and the signal to leakage noise ratio (SLNR) precoder in median to high mobility scenarios.