Intensive Tasks Offloading Decision Algorithms in Satellite Edge Computing Systems

To address the offloading allocation problem for intensive tasks in Satellite Edge Computing (SEC) systems, a Sequence-to-Sequence (Seq2Seq) and Deep Reinforcement Learning (DRL) based decision generation algorithm, enhancing resource allocation efficiency in SEC systems.

I. Introduction

Cloud computing centers are hubs for data processing and storage. They are facing huge challenges in terms of physical infrastructure and computing power due to increasing data traffic [1]. Mobile Edge Computing (MEC) shifts data processing tasks from centralized data centers to the edge of the network, thereby reducing data transmission distances, lowering latency, and improving service responsiveness. Satellite Edge Computing (SEC) systems combining satellite networks with edge computing, distributing data processing and storage functions globally [2].