Resource Orchestration in UAV-assisted NOMA Wireless Networks: A Labor Economics Perspective

The emergence of Unmanned Aerial Vehicles (UAVs) as part of the safety-critical and traffic alleviation infrastructure in 5G and beyond wireless networks, promotes the rethinking of the conventional resource orchestration management. In this paper, we propose a novel methodology that treats the uplink power allocation problem in UAV-assisted wireless networks, operated under Non-Orthogonal Multiple Access (NOMA), based on the principles of labor economics and Contract Theory (CT). The proposed approach specifically targets the challenge of imperfect Channel State Information (CSI) due to the uncertainties of the wireless links. The users are characterized by types that depend on their experienced channel conditions, which are typically unknown to the UAVs, while the latter probabilistically estimate the users' types. The users' transmission powers are iteratively optimized and determined, while an Reinforcement Learning (RL)-empowered user-to-UAV association procedure is realized. The overall framework is evaluated via modeling and simulation regarding its proper operation, effectiveness and efficiency, under different scenarios.