Energy Efficiency Maximization for Partially-Connected Hybrid Beamforming Architecture With Low-Resolution DACs
Hao Gao, Danpu Liu, Zhilong Zhang
Abstract
The combination of practical and easily deployable partially-connected (PC) hybrid beamforming (HBF) architecture and low-resolution (LR) quantizers guides a new direction for future millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) green communication. To fill the research gap in this area, we aim to propose an energy-efficient transmission scheme for the multi-user (MU) multiple-input multiple-output (MIMO) system based on LR digital-to-analog converters (DAC)-aided PC-HBF architecture. For this purpose, we formulate the energy efficiency (EE) maximization problem for jointly optimizing beamforming, power allocation, and DAC resolution selection. Confronted with a non-deterministic polynomial (NP)-hard mixed integer nonlinear programming problem, the original problem is decoupled into several subproblems with respect to each optimization variable, and we design targeted solutions respectively. More specifically, the beamforming design is primarily focused on mitigating spectral efficiency (SE) bottlenecks due to quantization distortion noise and MU interference, which distinguishes our work from the existing scheme. Besides, we propose a quadratic transformation (QT)-based power allocation scheme and a coordinate ascent search (CAS) for DAC resolution selection to maximize EE. Finally, we propose an innovative two-layer nested (TLN) scheme integrating the above design for a stable solution to the original problem. Simulation results show that the proposed TLN scheme has a fast convergence rate and significantly outperforms existing schemes in terms of EE and SE. In addition, a proposed low-complexity (LC) scheme exhibits performance close to that of the TLN scheme and demonstrates good robustness.