Machine learning methodology for TEC prediction using global positioning system signal measurements

Abstract Applications based on the modern Global Navigation Satellite Systems (GNSS) are primarily real-time applications that require high–precision positioning, navigation, and timing information. Propagation of Global Positioning System (GPS) signals through the ionosphere causes delays, allowing opportunities to develop the appropriate ionosphere/space weather nowcast and predict services. The propagation of Global Positioning System (GPS) signals through the ionosphere causes delays, allowing the development of nowcasting and forecasting model for the appropriate ionosphere/space weather services. In view of cosmic meteorological conditions are active (dynamic perturbations), observations of a single ionospheric attribute are not sufficient. Therefore, the accurate model is essential for nowcasting and predicting ionospheric perturbations, considering the effects of space meteorological conditions. Therefore, in this paper describes a new prediction approach, Extreme Kernel–Based Learning Machine (KELM), and introduces new relevance related to ionospheric activity services. The GPS–Total Electron Content (TEC) measurements of Bangalore location (13.02° E and 77.57° N) from 2009 to 2016 were considered for investigation. The outcomes of the proposed approach achieve good forecasting illustration at different seasons and solar period conditions. The error measurements MAE, MAPE, and RMSE, were 0.62, 8.29 and 0.81 TECU (KELM), and 1.44, 14.85, and 1.95 TECU (Auto Regressive Moving Average). With the ability, it would probably be implemented in real–time development applications.

[1]  J R K Kumar Dabbakuti,et al.  Modeling and analysis of ionospheric TEC variability from GPS–TEC measurements using SSA model during 24th solar cycle , 2021 .

[2]  J R K Kumar Dabbakuti,et al.  Characterization of ionospheric variability in TEC using EOF and wavelets over low-latitude GNSS stations , 2016 .

[3]  J R K Kumar Dabbakuti,et al.  Development of multivariate ionospheric TEC forecasting algorithm using linear time series model and ARMA over low-latitude GNSS station , 2019, Advances in Space Research.

[4]  S. Panda,et al.  Performance analysis of quiet and disturbed time ionospheric TEC responses from GPS-based observations, IGS-GIM, IRI-2016 and SPIM/IRI-Plas 2017 models over the low latitude Indian region , 2019, Advances in Space Research.

[5]  J R K Kumar Dabbakuti,et al.  Modeling of ionospheric TEC corrections based on modified mapping function and Gram–Schmidt orthogonalization at low latitude region , 2020 .

[6]  S. Panda,et al.  Analysis of ionospheric TEC from GPS, GIM and global ionosphere models during moderate, strong, and extreme geomagnetic storms over Indian region , 2019, Acta Astronautica.

[7]  Venkata Ratnam Devanaboyina,et al.  Analysis of local ionospheric variability based on SVD and MDS at low-latitude GNSS stations , 2016, Earth, Planets and Space.

[8]  D. Venkata Ratnam,et al.  Modeling and analysis of GPS-TEC low latitude climatology during the 24th solar cycle using empirical orthogonal functions , 2017 .

[9]  J. R. K. Kumar Dabbakuti,et al.  Modeling of Ionospheric Time Delays Based on a Multishell Spherical Harmonics Function Approach , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[10]  Venkata Ratnam Devanaboyina,et al.  Modeling of local ionospheric time varying characteristics based on singular value decomposition over low-latitude GPS stations , 2018, Astrophysics and Space Science.

[11]  J R K Kumar Dabbakuti,et al.  Ionospheric monitoring system based on the Internet of Things with ThingSpeak , 2019, Astrophysics and Space Science.

[12]  G. Sivavaraprasad,et al.  Short-term forecasting of ionospheric total electron content over a low-latitude global navigation satellite system station , 2017 .

[13]  Jyothi Ravi Kiran Kumar Dabbakuti,et al.  Application of Singular Spectrum Analysis Using Artificial Neural Networks in TEC Predictions for Ionospheric Space Weather , 2019, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[14]  J R K Kumar Dabbakuti,et al.  Ionospheric time delay corrections based on the extended single layer model over low latitude region , 2019, Geodesy and Geodynamics.

[15]  J R K Kumar Dabbakuti,et al.  Modeling of GPS-TEC using QR-decomposition over the low latitude sector during disturbed geomagnetic conditions , 2019, Advances in Space Research.

[16]  O. F. Althuwaynee,et al.  Ionospheric TEC from the Turkish Permanent GNSS Network (TPGN) and comparison with ARMA and IRI models , 2017 .

[17]  S. Panda,et al.  Mathematical modelling of ionospheric TEC from Turkish permanent GNSS Network (TPGN) observables during 2009–2017 and predictability of NeQuick and Kriging models , 2018 .

[18]  J R K Kumar Dabbakuti,et al.  Ionospheric forecasting model using fuzzy logic-based gradient descent method , 2017 .

[19]  Sivavaraprasad Gampala,et al.  Application of Synchrosqueezing Transform (SST) to Forecast Ionospheric Delays using GPS Observations , 2017 .

[20]  J R K Kumar Dabbakuti,et al.  Performance evaluation of linear time-series ionospheric Total Electron Content model over low latitude Indian GPS stations , 2017 .

[21]  D. Venkata Ratnam,et al.  Implementation of Hybrid Ionospheric TEC Forecasting Algorithm Using PCA-NN Method , 2019, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[22]  D. Venkata Ratnam,et al.  Improvement of Indian-Regional Klobuchar Ionospheric Model Parameters for Single-Frequency GNSS Users , 2018, IEEE Geoscience and Remote Sensing Letters.

[23]  N. Tripathi,et al.  Linear time series modeling of GPS-derived TEC observations over the Indo-Thailand region , 2018, Journal of Geodesy.

[24]  Kiran Kumar Anumandla,et al.  Implementation of storm‐time ionospheric forecasting algorithm using SSA–ANN model , 2020, IET Radar, Sonar & Navigation.