Research Article
Cloud Change Prediction System Based on Deep Learning
@INPROCEEDINGS{10.1007/978-3-030-69069-4_25, author={Dai Zheng and Zhao Kanglian and Li Wenfeng}, title={Cloud Change Prediction System Based on Deep Learning}, proceedings={Wireless and Satellite Systems. 11th EAI International Conference, WiSATS 2020, Nanjing, China, September 17-18, 2020, Proceedings, Part I}, proceedings_a={WISATS}, year={2021}, month={2}, keywords={Laser communication Cloud Prediction Network Cloud sequence prediction}, doi={10.1007/978-3-030-69069-4_25} }- Dai Zheng
Zhao Kanglian
Li Wenfeng
Year: 2021
Cloud Change Prediction System Based on Deep Learning
WISATS
Springer
DOI: 10.1007/978-3-030-69069-4_25
Abstract
In satellite-to-ground laser communications, the laser beam is susceptible to the effects of atmospheric media when it passes through the atmosphere. The main reason is that the laser beam will be absorbed and scattered by the cloud when it passes through the cloud, causing the communication link to be blocked. In order to know the cloud cluster information around the laser beam in advance, this paper proposes a Cloud Prediction Network (CloudNet) model, which classifies first, then predict the cloud trajectory for the next 100 s by collecting clouds images over a ground station, so as to reasonably allocating the resources of the link and select the ground stations. The experimental results show that the prediction accuracy of the model is up to 81% under the condition of 5% error.
