Hybrid Beamforming for Large Antenna Arrays with Phase Shifter Selection

This paper proposes an asymptotically optimal hybrid beamforming solution for large antenna arrays by exploiting the properties of the singular vectors of the channel matrix. It is shown that the elements of the channel matrix with Rayleigh fading follow a normal distribution when large antenna arrays are employed. The proposed beamforming algorithm is effective in both sparse and rich propagation environments, and is applicable for both point-to-point and multiuser scenarios. In addition, a closed-form expression and a lower-bound for the achievable rates are derived when analog and digital phase shifters are employed. It is shown that the performance of the hybrid beamformers using phase shifters with more than 2-bits resolution is comparable with analog phase shifting. A novel phase shifter selection scheme that reduces the power consumption at the phase shifter network is proposed when the wireless channel is modeled by Rayleigh fading. Using this selection scheme, the spectral efficiency can be increased as the power consumption in the phase shifter network reduces. Compared to the scenario that all of the phase shifters are in operation, the simulation results indicate that the spectral efficiency increases when up to 50% of phase shifters are turned off.