Intelligent Heterogeneous Aerial Edge Computing for Advanced 5G Access

In the context of the Internet of Things (IoT), aerial computing platforms (ACPs), such as unmanned aerial vehicles and high-altitude platforms with edge computing capabilities have the potential to significantly expand coverage, enhance performance, and handle complex computational tasks for IoT devices (IoTDs). Non-orthogonal multiple access (NOMA) has also emerged as a promising multiple access technology for advanced 5G networks. This paper presents a multi-ACP-enabled NOMA edge network, which enables heterogeneous ACPs to provide computational assistance to IoTDs. To minimize delay and energy consumption, we formulate a joint task offloading and resource allocation problem that considers IoTD association, offloading ratio, transmit power, and computational resource allocation variables. To address the complexity of the optimization problem, it is modeled as a multi-agent Markov decision process and solved using a multi-agent deep deterministic policy gradient (MADDPG)-based solution. Extensive simulation results demonstrate that the proposed MADDPG-based framework can remarkably adapt to the dynamic nature of multi-ACP-enabled NOMA edge networks. It consistently outperforms various benchmark schemes regarding energy efficiency and task processing delay across different simulated scenarios.