5:15 PM - 6:30 PM
[ACG43-P01] Machine learning for remote sounding of the ionosphere in Taiwan
Keywords:ionospheric sounding, machine learning
A. V. Dmitriev and NCU AI group
Global Navigation Satellite System (GNSS) plays crucial role for positioning and navigation. The GNSS accuracy is directly controlled by the conditions in the ionosphere. Routing monitoring of the ionospheric conditions is conducted by ground-based ionosondes, which provide experimental information about radio-echoes from ionospheric layers, so-called ionograms. The ionograms contain a lot of noise from numerous artificial transmitters such that the recovery of ionospheric signals becomes a very difficult problem. We present application of various machine learning (ML) techniques for determination of the signals from ionospheric E and F layers in very noisy environment of Taiwan, where the noise is comparable with the signal (SNR ~ 1). It is shown that deep ML techniques allow increasing of the determination accuracy up to 80% even for extremely nosy conditions.
NCU AI group: Chang Y.-C., Dmitriev A., Hsieh M.-C., Hsu H.-W., Huang G.-H., Li Y.-H., Lin C.-H., Lin Y.-C., Mendoza M., Tsai L.-C., Tsogtbaatar E.
Global Navigation Satellite System (GNSS) plays crucial role for positioning and navigation. The GNSS accuracy is directly controlled by the conditions in the ionosphere. Routing monitoring of the ionospheric conditions is conducted by ground-based ionosondes, which provide experimental information about radio-echoes from ionospheric layers, so-called ionograms. The ionograms contain a lot of noise from numerous artificial transmitters such that the recovery of ionospheric signals becomes a very difficult problem. We present application of various machine learning (ML) techniques for determination of the signals from ionospheric E and F layers in very noisy environment of Taiwan, where the noise is comparable with the signal (SNR ~ 1). It is shown that deep ML techniques allow increasing of the determination accuracy up to 80% even for extremely nosy conditions.
NCU AI group: Chang Y.-C., Dmitriev A., Hsieh M.-C., Hsu H.-W., Huang G.-H., Li Y.-H., Lin C.-H., Lin Y.-C., Mendoza M., Tsai L.-C., Tsogtbaatar E.