Low-complexity and fast-convergence linear precoding based on modified SOR for massive MIMO systems

Abstract

Linear zero forcing (ZF) precoding can obtain near-optimal performance as the number of base station (BS) antennas increases, but involve complex matrix inversion calculations whose computational complexity is cubic with respect to the number of user equips (UEs) in massive multiple-input multiple-output (MIMO) systems. In this paper, a low-complexity modified successive over-relaxation-based ZF (MSOR-ZF) linear precoding is proposed, in which complicated matrix inversion is directly avoided through the use of MSOR iteration. Moreover, to further increase the convergence rate, the proposed MSOR-ZF precoding exploits the diagonal dominant property of the matrix instead of the original zero vector solution to achieve a satisfactory performance with a reduced iteration number. The proposed precoding is superior in convergence rate and reduces the computational complexity by approximately one order of magnitude, which is indicated by theoretical analysis. Simulation results demonstrate that the proposed precoding can achieve better achievable sum-rate and bit-error-ratio (BER) performance than some existing iterative-based precoding schemes with lower complexity.