Work

Trajectory Planning and User Grouping for NOMA-Assisted UAV Data Collection

Year: 2026

Type: article

Abstract: In this work, we investigate a NOMA based data collection system for UAVs. Our goal is to minimize the collection task completion time by considering UAV’s 3D trajectory and user grouping. The problem is a mixed integer problem and the variables are coupled with each other, which makes it difficult to solve by traditional optimization methods. Therefore, we decompose the problem by first designing a deep reinforcement learning (DRL)-based method to solve the user grouping problem for a given time slot length and channel gain. Then, considering the limited computational capability of UAV, we design the distributed DRL algorithm to solve the problem of 3D trajectory planning and user grouping. The distributed DRL approach enables the complex training computations in DRL to be placed on the ground instead of on the UAV. Simulation results show that the proposed method can effectively reduce the task completion time in UAV data collection system.

Source: IEEE Open Journal of the Communications Society

DOI: 10.1109/ojcoms.2026.3707672