Resource Allocation for Multi-UAV and NOMA-enabled IoT Data Collection System

To improve the transmission performance of multiunmanned aerial vehicle-assisted, non-orthogonal multiple access (NOMA)-enabled Internet-of-things (IoT) data collection systems, we explore the joint optimization of user association, three-dimensional placement of UAV, decoding order, and power allocation. Our goal is to maximize the sum proportional data rate of all IoT devices using a probabilistic channel model. To cope with this mixed-integer nonconvex programming problem, we propose a resource allocation strategy based on an alternating optimization algorithm. This approach employs fractional programming theory, neurodynamic theory, and convex optimization theory to address the the sub-problems concerning each type of resource allocation variables. Simulation results show that the proposed strategy improves transmission performance by 9.60% compared with the existing K-means algorithm-based strategy. Moreover, it significantly reduces computation time by 99.33% compared with the global-optimal strategy, with only 0.26% compromise on optimized performance.