Optimizing UAV trajectories for quality of service enhancement in 5G urban deployments

The rapid proliferation of bandwidth-hungry applications and the limitations of fixed terrestrial infrastructure in urban environments have driven interest in deploying unmanned aerial vehicles (UAVs) as adaptive aerial base station (BS) within fifth generation (5G) wireless networks. In this work, we develop a three-dimensional (3D) simulation framework that incorporates 3GPP-standardized clustered delay line (CDL) channel models, probabilistic user mobility, 5G-compliant protocol stacks, and both single-input single-output (SISO) and multiple-input multiple-output (MIMO) transmission schemes in an urban macro cell (UMa) setting. We formulate a trajectory planning optimization problem that balances average signal-to-noise ratio (SNR) and its variability to derive flight paths under geographic constraints. Through extensive simulations, we evaluate multiple UAV mobility strategies including centroid following, maximum SNR expected value, minimum SNR standard deviation, weighted SNR trade-off, and fixed hovering, using throughput and block error rate (BLER) as the primary quality of service (QoS) performance indicators. Results show that the weighted SNR strategy achieves consistently high throughput with reduced variability, while its MIMO configuration yields an average throughput gain of 193% compared to SISO, highlighting the combined benefits of trajectory optimization and spatial diversity. These findings validate the proposed optimization criterion and underscore the substantial benefits of aerial mobility and spatial diversity in next-generation wireless networks.