A Review on Task Offloading and Resource Allocation in Aerial and Satellite-Assisted MEC Systems

This review provides a detailed analysis of recent advancements in resource allocation and task offloading strategies to enhance energy efficiency in aerial and satellite-assisted MEC systems. With the increasing demand for low-latency and high-computation services from IoT devices, integrating terrestrial, aerial, and satellite networks has emerged as a promising solution to expand coverage and boost computational capabilities. However, these integrated MEC systems face significant challenges, including dynamic task arrival rates, limited and heterogeneous resources, and fluctuating communication quality across diverse network layers. The review categorizes existing research into three key domains: aerial-assisted MEC, satellite-assisted MEC, and combined aerial and satellite-assisted MEC systems. It examines various optimization approaches, including deep rein-forcement learning (DRL), game theory, and hybrid algorithms, developed to address the complex problems of offloading and resource allocation in these environments. Furthermore, the review identifies open research challenges and potential future directions, such as advancing DRL techniques, to address current limitations. By offering insights into the state of the field, this review highlights research gaps and proposes pathways to improve the operational efficiency and energy management of aerial and satellite-assisted MEC systems.