Latency-Energy Efficient Task Offloading in the Satellite Network-Assisted Edge Computing via Deep Reinforcement Learning
Jian Zhou, Juewen Liang, Lu Zhao, Shaohua Wan, Hui Cai, Fu Xiao
Abstract
As the demand for global computing coverage continues to surge, satellite edge computing emerges as a pivotal technology for the next generation of networks. Unlike ground-based edge computing, Low Earth Orbit (LEO) satellites face distinctive challenges, including high-speed mobility and resource limitations, etc. Therefore, effectively utilizing LEO satellites for global coverage services is crucial but challenging due to their dynamic coverage areas and diverse task requirements. To address these challenges, we introduce a novel dual-cloud edge collaborative task offloading architecture in the satellite network-assisted edge computing environment, namely, <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</u>atellite-<underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">G</u>round <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</u>ask <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">O</u>ffloading (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SGTO</i>). The architecture employs a Geostationary Earth Orbit (GEO) satellite and a ground cloud computing center as satellite cloud and ground cloud, respectively, and LEO satellites as edge nodes. We formally define the task offloading problem in the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SGTO</i> with the aim of minimizing the average latency and average energy consumption. We then propose an adaptive approach named <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SGTO-A</i> from the perspective of satellites to adaptively solve the problem leveraging deep reinforcement learning. Specifically, we transform the task offloading problem into a Markov decision process and adopt the generalized proximal policy optimization (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GePPO</i>) algorithm to solve the problem. Finally, experimental results demonstrate that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SGTO</i> architecture and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SGTO-A</i> outperform the representative approaches in terms of average latency, average energy consumption and running time.