Performance Improvement in Large-Scale MU-MIMO System with Multiple Antennas on User Side in a Single-Cell Downlink System

Abstract

Large-scale multi-user multiple input multiple output (LS-MU-MIMO) is a promising technology to meet the demands for higher spectral and energy efficiency in the next-generation communication system. LS-MU-MIMO technology can provide 50 times or more network throughput by enhancing the spectral efficiency without cell densification and extra bandwidth requirements. This paper focuses on improvement in the spectral and energy efficiency for a single-cell LS-MU-MIMO system by considering multiple antennas at the user end with the linear precoding schemes. Spectral efficiency of zero-forcing (ZF), minimum mean square error (MMSE) and maximum ratio transmission (MRT) precoding schemes are derived, analyzed and compared for single- and double-antenna users with perfect and imperfect channel state information (CSI). MMSE precoding scheme shows the optimum spectral efficiency as compared to MRT and ZF precoding schemes for a LS-MU-MIMO system. We consider downlink and uplink transmission with various processing techniques at the base station and designed a system model for power consumption to evaluate the energy efficiency of the large MU-MIMO system. We derived energy efficiency expressions for maximal ratio, zero-forcing (ZF) and MMSE processing schemes. Numerical results reveal that the energy efficiency with double-antenna user equipments (UEs) has been enhanced as compared to the UEs having single antenna. Simulation results show that the proposal of adding multiple antennas at the user end can provide a substantial improvement in spectral and energy efficiency as compared to single-antenna users for perfect and imperfect CSI with various precoding schemes.