User-Centric Joint Beamforming for Multi-Antenna UAV Networks With 3D Blockage Effects

In unmanned aerial vehicle (UAV) networks, introducing multiple-input/multiple-output (MIMO) technology brings promising capacity gains by exploiting spatial multiplexing. However, for UAV communications in urban scenarios, dense buildings block air-to-ground links, which leads to non-consistent service for users. In this paper, a user-centric joint beamforming design for multi-UAV networks aiming at maximizing the sum-rate performance is proposed, where the distributed antennas are deployed in UAV clusters forming virtual MIMO links to seamlessly guarantee user service under the 3D blockage effect. Specifically, the UAV clusters are constructed dynamically in a user-centric way for cooperative transmissions, which are dominated by the line-of-sight (LoS) connections considering building blockage effects. In addition, the selected user-to-UAV links must be less affected by the building blockages which are measured through the established channel model from comprehensive dimensions including sizes, locations, and heights. Within the UAV clusters, the proposed joint beamforming is performed based on the majorization-minimization (MM) algorithm for the NP-hard problem under per-antenna power constraints, and the optimal UAV parameters including UAV placement and antenna configuration with a fixed total number of antennas are investigated under different blockage effect. Simulation results demonstrate that the proposed design can improve sum-rate performance by more than
$1\times $
compared with single UAV deployment under densely-located building environments.