Aerial Computing Offloading by Distributed Deep Learning in Collaborative Satellite-terrestrial Networks
Haofei Li, Chen Chen, Cong Li, Lei Liu, Guan Gui
Abstract
With the rapid development of the Internet of Things (IoT) and its derivative technologies such as Aerial IoT (AIoT), the number of smart devices and their computing requirements have exploded. It makes it difficult for the existing ground base stations to guarantee the user's computing requirements and low latency requirements. The edge computing architecture of orbiting satellites, which provides feasible solutions to the above challenges. In this paper, we design a computing offloading strategy for users to minimize the weighted sum of latency and energy consumption in the cloud-edge-ground three layers hybrid AIoT architecture. Tasks can be calculated locally, on the satellite side, or forwarded to the cloud computing center. Then, the collaborative satellite-terrestrial network distributed offloading algorithm(CNDO) is proposed based on parallel neural networks to give multiple optimal offloading decisions in AIoT. Experimental results show that this scheme has a better performance compared with other comparison schemes.