Table 1

Summary of the described ionospheric parameters forecasting based on NN.

Reference Approach Prediction lead time Main features
Sai Gowtam & Tulasi Ram (2017) ANN prediction model of NmF2 and hmF2 as function of LT, LAT, LONG, SEASON Climatological model NmF2 and hmF2 percentage error ranged between 15% and 20%. The NmF2 model performance decreases at low latitudes during the predawn hours and around midnight at middle-high latitude.
The hmF2 model performance decreases at low latitude at postsunset.
Tulunay et al. (2006) Regional TEC forecasting model, NN 10 min up to 1 h Short term forecasting. RMSE increases with forecasting horizon up to about 4 TECU over mid latitude grid points.
Habarulema et al. (2011) ANN regional TEC variability forecasting model as function of time of the day, season, solar and magnetic activity, latitude and longitude Climatological model The model accuracy is decreases under geomagnetically disturbed conditions.
Huang & Yuan (2014) TEC single station forecasting model based on BP NN and RBF NN 30 mins Short term forecasting. RMSE less than 5 TECU, the model performance decreases at lower latitude.
Cherrier et al. (2017) Global TEC maps forecasting model based on recurrent NN and TEC global map by CODE 2–48 h RMSE increases with forecasting horizon. The model performance decreases under disturbed conditions.
Perez (2019) Global TEC maps forecasting model NN combined with GIM map 24-h or more Equatorial Ionospheric Anomaly not represented

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.