Joint Optimization of Beamforming and 3D Array Steering for Multi-Antenna UAV Communications

In this paper, we consider unmanned aerial vehicle (UAV)-aided downlink communication using a rotatable array of directional antennas such as half-wavelength dipoles. The antenna array is mounted onboard the UAV using a gimbal and can be flexibly rotated in the three-dimensional (3D) space. As such, the directional gain pattern of dipoles and the array beamforming can be both best exploited to facilitate efficient information transmission to multiple low-priority (or secondary) users while mitigating co-channel interference for another high-priority (or primary) user coexisting with but not served by the UAV. Assuming that the direction of the high-priority user is known, we jointly optimize the electrical beamforming and mechanical steering of the rotatable dipole array for maximizing the weighted sum-rate achievable by the low-priority users while minimizing the interference power radiated in the direction of the high-priority user. The formulated optimization problem is nonconvex and generally intractable. Exploiting its special problem structure, we decompose the problem into several convex and manifold optimization subproblems and further propose a low-complexity iterative algorithm based on proximal block coordinate descent for solution. Our simulation results verify the convergence of the proposed algorithm. Moreover, compared with systems employing non-rotatable and rotatable arrays of isotropic antennas, the rotatable dipole array can flexibly adjust the gain patterns in different azimuth and elevation angles to increase the communication throughput by up to 300% and 77%, respectively.