一种基于SSOR的大规模MIMO线性预编码方案

A Linear Precoding Scheme for Massive MIMO Systems Based on SSOR Method

大规模多输入多输出系统(Massive MIMO)由于天线数和用户数太大,导致预编码矩阵在求逆是复杂度过高。为了降低复杂度,本文提出了一种基于对称逐步超松弛(SSOR)的线性预编码方案,相比传统的规则化迫零(RZF)预编码方案,本文所提的方案在没有任何性能损失的情况下可以降低一个量级的运算复杂度,为了保证所提SSOR预编码方案的性能,提出一种仅依靠天线配置的简单的量化松弛参数。

Massive multiple-input multiple-output（MIMO） has a large number of attennas and users, which leads to high computational complexity in precoding matrix.To solve this problem,this paper proposed a linear precoding scheme based on symmetric successive overrelaxation（SSOR）. Compared to conventional RZF scheme,it can reduce one order of magnitude with negligible performace loss. To guarantee the performance of SSOR-based precoding, we also proposed a simple quantified relaxation parameter for SSOR-based scheme, which only depends on the MIMO systemconfiguration.