Machine Learning-Based Multi-UAVs Deployment for Uplink Traffic Sizing and Offloading in Cellular Networks

Traffic offloading in cellular networks is considered an evolving application of unmanned aerial vehicles (UAVs). UAVs have attractive characteristics for this application, such as the ease deployment, relatively low cost and line-of-sight signal propagation. This paper proposes a machine learning-based deployment temporary base stations (BSs) to complement communication systems times excess traffic loads. In role, UAV tasked with proper sizing mixed demands on terrestrial BSs subsequent traffic, given its different QoS requirements. We achieve objective by optimizing number needed their three-dimensional (3D) positions. A estimation technique based Autoregressive Integrated Moving Average (ARIMA) model utilized estimate demand. Our proposed machine-learning approach, reinforcement learning (RL) methodology, aims obtain real-time results close solution’s optimal bound. Simulation show that RL solution achieves close-to-optimal objectives. The approach also shown clearly outperform commonly used generic situations.