Performance of Regularized Zero Forcing (RZF) Precoding for Multiuser Massive MIMO-GFDM System over mmWave Channel

Abstract

To achieve its goal of providing higher communication capacity, 5G technology needs larger bandwidth availability, which is quite hard to be met in middle and low frequency bands. millimeter Wave (mmWave) has emerged as a solution for bandwidth scarcity on microwave bands frequencies. As one of its property is the short wavelength, mmWave is suitable for massive MIMO. Even though mmWave experience higher path loss, massive MIMO system can provide sufficient gains to compensate the serious signal attenuation by using the precoding technique. In this paper, we evaluate the performance of multi-user massive MIMO system over the mmWave channel. The geometric Saleh-Valenzuela (S-V) model is employed to characterize small scale fading of mmWave massive MIMO channel. To further boost system performance and capacity of 5G communication system, we combine multi-user massive MIMO technique to GFDM waveform. As one of the new waveform candidates for 5G, previous studies suggest that GFDM displays better performance compared to OFDM. RZF precoding is employed to the transmitter side of a multi-user massive MIMO-GFDM system. BER performance are evaluated with different configurations of GFDM system and antenna parameters. Simulation results show that configuration with smaller roll-off factor (γ) achieves lowest BER. The combination of number of subcarrier and sub-symbol also contributes in BER performance. Higher number of BS antenna provides better BER value under low SNR. Overall, among the proposed configuration, massive MIMO-GFDM with K×M = 64×4 and γ = 0.1 outperforms others.