Data Collection Mechanism for UAV-Assisted Cellular Network Based on PPO

• DOI: 10.3390/electronics12061376
• Publication Date: 2023-03-14T06:33:36Z
• Authors: Tuo Chen; Feihong Dong; Hu Ye; Yun Wang; Bin Wu
• Keywords: Drone

Unmanned aerial vehicles (UAVs) are increasingly gaining in application value many fields because of their low cost, small size, high mobility and other advantages. In the scenario traditional cellular networks, UAVs can be used as a kind mobile base station to collect information edge users time. Therefore, provide promising communication tool for computing. However, due limited battery capacity, these may not able completely all information. The path planning ensure that UAV collects much data possible under flight distance, so it is very important study UAV. addition, particularity air-to-ground communication, flying altitude have crucial impact on channel quality between user. As mature technology, deep reinforcement learning (DRL) an algorithm field machine which deployed unknown environments. Deep applied collection UAV-assisted find best height joint optimization scheme, ensures more condition energy consumption, save human material resources possible, finally achieve higher value. this work, we transform problem into Markov decision process (MDP) problem. By applying proximal policy (PPO) algorithm, our proposed realizes adaptive Simulations conducted verify performance scheme compared conventional scheme.