Prediction of Ionospheric Irregularities using a Combination of Machine Learning Algorithms

Ionospheric irregularities can be caused by from sun activity variations, which may cause irregularities in electron density within the ionospheric layer and, subsequently, plasma perturbations. Typical examples of these irregularities are ionospheric scintillations. The ionospheric irregularities can cause fluctuations in the signal intensity transmitted from the satellite by reducing the signal-to-noise ratio. In addition, scintillation can lead to extreme fluctuations in the phase of a signal transmitted. Ionospheric irregularities originate destructive effects on radio signals transmitted from global navigation satellite systems (GNSS). This phenomenon can generate fluctuations in the signal intensity transmitted from the satellite by decreasing the signal-to-noise ratio of the transmitted wave. The primary purpose of this research will be to detect, model, and predict ionospheric irregularities using a hybrid machine learning algorithm. In addition, using prediction values obtained from the proposed Hybrid models allow measuring the effect of ionospheric perturbations on GNSS ground-based precise positioning accuracy. This modeling and prediction algorithm can contribute to reducing the error of the ionospheric irregularities for satellite-based communication and navigation systems performance. For this purpose, near the equatorial ionization anomaly (EIA), GNSS ground-based stations in South America, are recommended since ionospheric disturbances most impact these regions. The proposed method can play a precaution role in alerting GNSS users that the observation epoch will be disturbed by ionospheric perturbations, and GNSS users can eliminate error-infected observations from the dataset.