QoE and cost-aware resource and interference management in aerial-terrestrial networks for vehicular applications

Abstract—In this paper, we address the deployment of Unmanned Aerial Vehicles (UAVs) as Unmanned Aerial Base Stations (UABSs) which cooperate with Macro Base Stations (MBSs) in an urban environment to serve vehicles, denoted as Ground User Equipments (GUEs), implementing vehicle-to-everything (V2X) services. As vehicles perform extended sensing, exchanging data with nearby GUEs through UAVs and MBSs links, we propose an Integer Linear Programming (ILP) model that jointly optimizes radio resources allocation and beamforming, while accounting for vehicular application requirements, backhaul capacity limits and interference between GUE-UABS and GUEMBS links. The model allows also to find a trade-off between benefits and cost of UABSs activation. Two system architectures are considered: a distributed model, where MBSs independently run the Radio Resource Management (RRM) algorithm sharing information with each other, and a centralized model, where MBSs send information to the network core, where the optimization algorithm runs. The study investigates interference through two resource allocation approaches, considering splitting and sharing of resources among UABSs and MBSs. Numerical evaluations demonstrate the effectiveness of using UABSs to improve the Quality of Experience (QoE) of GUEs. We also compare the two architectures, considering both resource pool assignments, and highlighting the impact of varying UABSs parameters and activation costs.