Enhancing performance of massive MIMO communication systems using iterative regularized zero-forcing precoding

Massive multiple-input multiple-output (massive MIMO) has become a key technology in modern cellular networks because it can improve spectral efficiency, data rate, and coverage. These gains, however, require efficient precoding algorithms. Conventional precoding schemes such as regularized zero-forcing (RZF) precoding offer a useful trade-off between interference suppression and noise amplification, but their direct matrix inversion becomes computationally expensive when the number of base-station antennas is large. In this paper, we propose an iterative regularized zero-forcing (IRZF) precoding method that replaces direct inversion with a Richardson-type iterative algorithm to perform precoding with low complexity. We have compared the performance of the proposed method with conventional Gauss-Seidel (GS), Jacobi (JA), Joint Jacobi and Gauss-Seidel (Jac-PCG), Truncated polynomial expansion (TPE), and Regularized zero-forcing (RZF) in terms of achievable rate, spectral efficiency, computational complexity, and energy efficiency. The results show that the proposed IRZF achieves high performance with low complexity. Therefore, IRZF is most attractive in operating regimes where the required number of iterations is moderate and the regularized Gram matrix is well-conditioned.