Multi-UAV Deployment and Resource Allocation for Energy Efficiency in NOMA Networks

This paper investigates the efficient deployment and resource allocation of multiple unmanned aerial vehicles (UAVs) to be integrated with wireless terrestrial networks as aerial base stations (BSs) and serve ground user devices by utilizing non-orthogonal multiple access (NOMA) scheme. Aiming to optimize the energy efficiency (EE) of ground BSs (GBSs) and deployed UAVs, a constrained EE maximization problem is formulated. To solve the optimization problem, we decompose the problem into three subproblems. Firstly, GBS association and subchannel assignment subproblem is formulated and solved by applying matching theory. Then, UAV deployment and subchannel assignment subproblem is formulated. To solve the problem, we propose a clustering strategy based on K-means algorithm to divide users into approximately equal-size clusters. Then, a dynamic many-to-many matching algorithm is proposed to assign subchannels to the users. Finally, a non-convex power allocation subproblem is formulated and solved iteratively using a two-stage quadratic transform. The numerical results demonstrate the effectiveness of our proposed scheme.