Modeling and prediction of ionospheric total electron content by time series analysis

Precise modeling and accurate prediction for the ionospheric total electron content(TEC) are crucial and remain as a challenge for GPS positioning and navigation, space weather forecast, as well as many other Earth Observation System(EOS). This research develops and analyzes a new prediction technique for the regional ionospheric TEC, based on time series analysis theory using autoregressive model (AR) to perform short-term ionospheric TEC prediction. The predicted TEC were then compared with the TEC measured by IGS, and with TEC from the International Reference Ionosphere(IRI) to assess the performance of the model. Preliminary results show that AR model could well describe the variation trend of the regional ionospheric TEC and has a good short-term performance of the ionospheric TEC prediction. The forecasting methodology based on the time series for the regional ionospheric TEC prediction is feasible.