An improved AOR-based precoding for massive MIMO systems

Abstract

Compared with traditional multiple-input-multiple-output (MIMO) systems, the number of base station (BS) antennas increased in massive MIMO systems. However, due to the huge number of antennas, linear precoding schemes are able to achieve the near-optimal performance. Conventional linear precoding schemes such as regularized zero forcing (RZF) precoding need to calculate the matrix inversion of large size which leads to high computational complexity. Although utilizing iterative algorithm approximate instead of matrix inversion can reduce the complexity, it leads to slow convergence or general bit error rate (BER) performance. To solve this problem, we proposes a linear precoding scheme based on the accelerated over relaxation (AOR) method. Moreover, we propose a simple way to choose the optimal accelerate factor so that it is only related to the system parameters and more suitable for practical application. Simulation results prove that the improved AOR-based precoding could convergence faster and had better performance of bit error rate (BER).