Hybrid Beamforming with Reduced RF Chain Based on PZF and PD-NOMA in mmWave Massive MIMO Systems

Abstract

Hybrid beamforming (HB) is a solution for reducing radio frequency (RF) chains in millimeter-wave (mmWave) massive MIMO systems. By reducing RF chains, HB can achieve energy efficient transmission compared to fully digital beamforming of massive MIMO system. However, if the resolution of RF precoder is not quantized, the energy consumption of HB may become similar to fully digital beamforming. And, if the resolution of RF precoder is heavily quantized, the spectral efficiency of HB is decreased significantly. Non-orthogonal multiple access (NOMA) is one of the emerging technologies for the fifth generation (5G) or beyond fifth-generation (B5G) wireless communication to support high spectral efficiency. Especially, power domain NOMA (PD-NOMA) can share same resources to transmit signals and divide the signals in power domain by using successive interference cancellation (SIC). In this paper, the adoption of NOMA shows increase of spectral efficiency in a low-resolution RF precoder. To apply NOMA, two users with high channel correlation are selected according to the proposed criterion. Then, the RF precoder and digital precoder that are applied to both users at the same time are determined. Finally, intra-pair user fairness is achieved by power allocation. The ratio of channel power is compared for each user in each pair. Then, users with high channel power are assigned with low power. It is shown that the spectral efficiency of the proposed scheme is higher than the conventional scheme in heavily quantized RF precoder.