Completion Time Minimization for UAV-Aided Communications with Rotatable Dipole Array

This paper investigates UAV-aided wireless communications with multiple users while utilizing a rotatable dipole array at the UAV. We jointly optimize user scheduling, array steering, and beamforming to minimize the UAV's mission completion time, i.e., the time required to deliver a specified data volume from the UAV to each user. To tackle the resulting nonconvex mixed-integer nonlinear program (MINLP) problem, we identify a hidden convexity in the optimization of continuous variables for given values of the discrete variables. Leveraging this result, we reformulate the original problem as a multi-stage dynamic programming (DP) problem and characterize its optimal solution via the Bellman optimality equation. We further propose a novel low-complexity one-step lookahead rollout (OSLR) algorithm based on approximate DP and semidefinite programming (SDP) jointly to optimize the discrete and continuous variables, respectively. Simulation results demonstrate that our proposed algorithm achieves significant reductions in the UAV's mission completion time compared to two baseline schemes, even when only a small number of dipoles are deployed at the UAV.