Reviews of recent progress on low-complexity linear detection via iterative algorithms for massive MIMO systems

Abstract

Massive multiple-input multiple-output (M-MIMO) can significantly enhance the spectrum efficiency of cellular networks by deploying hundreds of active elements at the base stations and is envisaged to become the key technology in 5th generation (5G) cellular networks. However, the large number of antennas required brings about tremendous challenges for practical implementation, especially for separation of the multiplexed data. Iterative approaches, such as Jacobi, Richardson, Gauss-Seidel (GS), successive overrelaxation (SOR), and symmetric successive overrelaxation (SSOR) have received great attention recently due to their low-complexity and high performance for signal detection. In this work, we provide a comprehensive review of recent progress in iterative based signal detection for massive MIMO systems. The system model of an iterative method based minimum mean square error (MMSE) signal detection is provided. The convergence behavior and complexity of the iterative approach based detectors are analyzed. Numerical results show that the iterative algorithm-based detectors can achieve a performance close to the classical MMSE detector with significantly less computational complexity.