A Novel 3D Beam Training Strategy for mmWave UAV Communications

The beam training technology which can overcome the easily-occurred beam misalignments is required urgently in unmanned aerial vehicles (UAVs) based millimeter wave (mmWave) communication systems. In this paper, a novel three-dimensional (3D) beam training strategy for UAV-assisted mmWave communications is proposed. The inverse discrete-space Fourier transform is introduced to construct the training beam with flat-topped characteristic. In addition, the hybrid beamforming (BF) system is taken into account and the greedy geometric (GG) algorithm is adopted to obtain the optimal beam. Numerical simulations are conducted to evaluate the performance and the simulation results demonstrate that our proposed 3D training strategy can provide both precise beams and high training efficiency for the 3D mmWave UAV communications.