Joint 3D Codebook Design and Beam Training for UAV Millimeter-Wave Communications

Owing to its excellent flexibility, unmanned aerial vehicles (UAV) have been widely adopted as aerial access points to provide data collection services for the Internet of Things (IoT) devices. Moreover, Millimeter-Wave (MmWave) aided UAV communications to achieve extremely high data rate has also become the hot issue in recent research. Codebook designing and beam training are the enabling technologies of MmWave communications. However, the existing solutions for communication systems cannot be directly applied in the UAV Mmwave communications, because of the increased complexity in moving and 3D UAV scenarios. Therefore, this paper focuses on joint codebook design and beam training. Firstly, a 3D codebook is designed which can provide flexible access for IoT devices and achieve the optimal system throughput. Then, based on the designed codebook, a Angle Forecast based Fast Beam Alignment (AFFBA) mechanism is proposed. This mechanism infers the potential angle range of ideal AoD from the beam adopted by the current serving UAV. Combing the angle range and the trajectory model of UAV, the optimal beam is forecasted. The proposed joint 3D codebook design and beam training significantly reduce the dimension of beam sweeping space. Simulation results demonstrate the superior performance of the proposed mechanism, and show that the proposed mechanism significantly reduce the beam sweeping space and effectively improve the normalized spectral efficiency (NSE) compared to existing exhaustive training mechanism.