On the Beamforming Design of Millimeter Wave UAV Networks: Power vs. Capacity Trade-Offs

The millimeter wave (mmWave) technology enables unmanned aerial vehicles (UAVs) to offer broadband high-speed wireless connectivity in fifth generation (5G) and beyond (6G) networks. However, the limited footprint of a single UAV implementing analog beamforming (ABF) requires multiple aerial stations to operate in swarms to provide ubiquitous network coverage, thereby posing serious constraints in terms of battery power consumption and swarm management. A possible remedy is to investigate the concept of hybrid beamforming (HBF) transceivers, which use a combination of analog beamformers as a solution to achieve higher flexibility in the beamforming design. This approach permits multiple ground users to be served simultaneously by the same UAV station, despite involving higher energy consumption in the radio frequency (RF) domain than its ABF counterpart. This paper presents a tractable stochastic analysis to characterize the downlink ergodic capacity and power consumption of UAV mmWave networks in an urban scenario, focusing on the trade-off between ABF and HBF architectures. A multi-beam coverage model is derived as a function of several UAV-specific parameters, including the number of UAVs, the deployment altitude, the antenna configuration, and the beamforming design. Our results, validated by simulation, show that, while ABF achieves better ergodic capacity at high altitudes, an HBF configuration with multiple beams, despite the use of more power-hungry RF blocks, consumes less power all the time with limited capacity degradation.