The purpose of this study is to verify the rationality and validity of directly adopting segmental modeling prediction of single-station regional scatter data by considering space environment information comprehensively. First, a single-station regional ionospheric model is constructed and predicted by the long short-term memory neural network (LSTM) method based on the single-station global positioning system total electron content (GPS-TEC) data of different regions (low-, mid-, and high latitude regions) and the space environment data. Then, the prediction results are compared and analyzed with the international reference ionosphere 2016 (IRI2016) model, CMONOC Regional Ionosphere Maps (RIM) data, and GPS measurement data. The results show that: i) the LSTM model forecasts are consistent with GPS-TEC observations at high, mid and low latitudes, and the forecast error is less than 3 TECu. The forecast accuracy is much better than that of the IRI2016 model and RIM TEC, and is less susceptible to anomalies. Geographically, the forecast MAE and RMSE of LSTM model decreases with increasing latitude. Among them, the relative accuracy of LSTM forecasts in low and mid latitudes is high, up to 82% or more; ii) the RIM data as a whole are more consistent with the measured data, but the RIM TEC is overestimated during the daytime, a phenomenon related to the discrete anomalies; iii) the IRI2016 model only captures the general trend of TEC. The IRI model forecast values are poor in daytime forecasting, and its overestimation becomes more obvious as the latitude increases, while the forecast performance is better in the evening. This study is a foundation for subsequent regional modeling and forecasting of the ionosphere, and can provide ionospheric constraints to support navigation positioning.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A Single-Station Ionospheric Forecast Model with LSTM Considering Multiple Factors


    Weitere Titelangaben:

    Lect. Notes Electrical Eng.


    Beteiligte:
    Yang, Changfeng (Herausgeber:in) / Xie, Jun (Herausgeber:in) / Xie, Ting (Autor:in) / Dai, Zhiqiang (Autor:in) / Zhu, Xiangwei (Autor:in)


    Erscheinungsdatum :

    2022-05-05


    Format / Umfang :

    11 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    A Single-Station Ionospheric Forecast Model with LSTM Considering Multiple Factors

    Xie, Ting / Dai, Zhiqiang / Zhu, Xiangwei | TIBKAT | 2022


    Ionospheric specification and forecast modeling

    Schunk, R. / Scherliess, L. / Sojika, J. | AIAA | 2001


    Ionospheric Specification and Forecast Modeling

    Robert W. Schunk / Ludger Scherliess / Jan J. Sojka | AIAA | 2002


    Forecast of global ionospheric TEC using an improved transformer model

    Wu, Xuequn / Fan, Cihang / Tang, Jun et al. | Elsevier | 2024