Comparative Analysis of Low Complexity Signal detection methods in Uplink Massive MIMO

Abstract

Massive multiple-input multiple-output (MIMO) is a promising technology of fifth-generation (5G) wireless communication systems to support performance gain, high spectral efficiency, and energy efficiency. Symbol detection is a computationally complicated task for a massive MIMO baseband receiver. It is not an ordinary job to design M-MIMO detector for a large number of antennas. In this study, we analyze the approximate message passing (AMP) algorithm along with approximate matrix inversion-based method detectors namely, Neumann series (NS), Symmetric Successive over-relaxation (SSOR), Jacobi method (JA), and Gauss-Seidel (GS), Numerical results present that the AMP algorithm outperforms well in the high configuration of antennas, furthermore as compared to AMP, the approximate inversion detectors (AID) method GS, SSOR are also outperformed well in the chosen configuration of antennas in M-MIMO. On the other hand, results also depict that the AMP algorithm detector having a higher convergence rate and less processing time compared to NS, SSOR, GS, and JA. We also present the Computational complexity of discussed methods. These results provide an important guideline for very-large-scale integration (VLSI) designers to choose a suitable M-MIMO detection algorithm according to the system requirement.