The Doppler frequency offset which caused by the high dynamic characteristics of LEO satellites increases the difficulty of signal recovery at the receiving terminal. Aiming at the Doppler frequency offset problem of LEO satellite communication systems, this paper proposes a Doppler frequency offset pre-compensation algorithm based on Multimodal Long-Short Term Memory (MLSTM-DPC). By judging the difference between the current ephemeris data and the current time, select single or multiple LSTM models to predict the orbital parameters. The predicted orbit parameters and traditional algorithms are used to extrapolate the orbit. Finally, the value of Doppler frequency offset pre-compensation is predicted. The simulation results show that the effective frequency offset ratio of MLSTM-DPC algorithm is 36.39% higher than that of comparison algorithm, and the computational time is significantly reduced. In a certain range, if the TLE data is farther from the reference time, the advantage of using MLSTM-DPC Doppler frequency offset precompensation algorithm for precompensation is more obvious. The algorithm provides a reliable pre-compensation mechanism for terminal fast synchronous access.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Multimodal LSTM forecasting for LEO Satellite Communication Terminal access


    Beteiligte:
    Li, Hongguang (Autor:in) / Liu, Yaoqi (Autor:in) / Shi, Jinglin (Autor:in) / Zhou, Yiqing (Autor:in) / Zhuo, Ruilian (Autor:in) / Li, Shaoyang (Autor:in)


    Erscheinungsdatum :

    2023-06-01


    Format / Umfang :

    2479379 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Business forecasting for communication satellite systems

    BAER, J. / EARLY, L. / KUMINS, L. | AIAA | 1966



    Attention Based Graph Bi-LSTM Networks for Traffic Forecasting

    Zhao, Han / Yang, Huan / Wang, Yu et al. | IEEE | 2020


    Forecasting total electron content (TEC) using CEEMDAN LSTM model

    Shaikh, Muhammad Muneeb / Butt, Rizwan A. / Khawaja, Attaullah | Elsevier | 2022