@inproceedings{094b3b2011bf44d283ff13799f9133e7,
title = "Satellite Dynamic Channel Prediction Based on LSTM Network",
abstract = "Satellite communication channel information are crucial for effective information transmission from ground stations to satellites. Current methods generally rely on the downlink to get the estimated channel conditions of satellites, however they are inapplicable in the event of an abrupt satellite downlink disconnect. Therefore, we propose a method for predicting the long-time carrier-to-noise ratio (CNR) of satellite communication channels. This method is based on the long and short term memory (LSTM) network model. By introducing the predicted position of the satellite, the accuracy of the CNR prediction is promoted. The simulation results show that the residuals of the predicted CNR within 1000 seconds are less than 4 dBHz, and the root mean square error(RMSE) is 2.7372 dBHz.",
keywords = "LSTM, channel prediction, deep learning, orbit prediction, satellite communication",
author = "Yongqing Wang and Yidan Wang and Yuyao Shen",
note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 7th IEEE Information Technology and Mechatronics Engineering Conference, ITOEC 2023 ; Conference date: 15-09-2023 Through 17-09-2023",
year = "2023",
doi = "10.1109/ITOEC57671.2023.10291710",
language = "English",
series = "ITOEC 2023 - IEEE 7th Information Technology and Mechatronics Engineering Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "150--155",
editor = "Bing Xu and Kefen Mou",
booktitle = "ITOEC 2023 - IEEE 7th Information Technology and Mechatronics Engineering Conference",
address = "United States",
}