A Low-Altitude Network Base Station Planning Model Based on PPO Algorithm

The rapid development of low-altitude unmanned aerial vehicles (UAVs) has led to significant communication demands. Leveraging cellular networks to support low-altitude UAV communication offers cost advantages. However, existing cellular networks suffer from inadequate aerial coverage and the complexity of urban terrestrial channels. To address these challenges, we propose a novel low-altitude network base station planning model based on the Proximal Policy Optimization (PPO) algorithm. Our approach involves calculating the low-altitude coverage capabilities of different base station types using ray tracing techniques. We construct a decision model for base station placement and utilize the PPO algorithm to determine optimal base station configurations. Through reinforcement learning training and simulation validation, we demonstrate that our model effectively generates base station deployment plans, achieving predefined signal coverage objectives.