Neural Network-Based Single-Station Ionospheric TEC Prediction at High Latitude

Total electron content (TEC) is one of the most important ionospheric parameters since the satellite era reflecting significant characteristic of ionosphere which has practical importance for its effects on radio propagation to earth. In this paper, a deep learning-based TEC prediction model at high latitude station is proposed, and a long short-term memory (LSTM) network is used to realize the short-term prediction of TEC at high-latitude station (79.9°N, 85.9°W). The short-term prediction performance of the BP neural network and the LSTM neural network for ionospheric TEC has been made comparison with the TEC measurements from 2010 to 2023. The results show that the root mean-square error (RMSE) of the LSTM model for TEC prediction in the next hour is less than 1.5 TECU, and the accuracy is 13% higher than that of the BP neural network at the optimal number of iterations.