Coordination des ressources radio dans les réseaux multi-UAV directionnels avec réutilisation spectrale

Abstract

Unmanned Aerial Vehicles (UAVs) have recently emerged as a promising solution for enhancing the coverage and capacity of next-generation wireless networks. However, in multi-UAV environments operating over a shared frequency band, radio resource reuse, directional beam steering, and transmit power control create strong interdependencies among communication links, making interference management particularly challenging. Although radio resource coordination has been widely investigated in wireless networks, existing approaches often rely on simplified assumptions or partially decoupled designs, which limit their ability to explicitly represent the coupling between beam steering, resource allocation, power control, and interference in directional multi-UAV scenarios. This work studies radio coordination in a directional multi-UAV downlink network within a quasi-static framework, where UAV positions are assumed to be given. A system model integrating three-dimensional propagation, directional antenna gain, structured spectrum reuse of radio resources, and SINR evaluation is first developed. To address this problem in the considered scenario, a block-based radio coordination procedure with constraint-aware acceptance is proposed. The procedure successively updates beam steering, radio resource allocation, and transmit power decisions. At each step, a candidate configuration is evaluated based on the SINR, the problem constraints, and the total throughput. The candidate configuration is retained only if it satisfies the constraints and improves or preserves the current performance. The objective is therefore to obtain a feasible and non-degrading radio configuration in a controlled scenario, without claiming global optimality for the general problem. Numerical results obtained in a controlled scenario suggest improved effective inter-UAV interference management and increased overall throughput compared with partially decoupled strategies. The observed iterative behavior shows a progressive reduction of throughput variations in the considered scenario, reflecting numerical stabilization of the procedure without constituting a proof of global optimality.

Date	14 Jul 2026
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Kim Khoa Nguyen (Supervisor) & Mohamed Cheriet (Co-supervisor)