On the predictability of foF2 twenty-four hour ahead using a support vector machine technique

This paper proposes a method for forecasting the ionospheric critical frequency, f₀F₂, 24 hour in advance using the support vector machine (SVM) approach. The inputs to the SVM network are the time of day, seasonal information, a 2 month running mean sunspot number (R2), a 3 day running mean of the 3 hour planetary magnetic Ap index, the solar zenith angle, the present value f_oF₂(t), the observation of f₀F₂ at t-23 time, and the previous 30 day running mean of f₀F₂ at t-23 time f_mF₂ (t-23). The output is the predicted f₀F₂ one hour ahead. The network is trained to use the ionospheric sounding data at Guangzhou, Changchun, Manzhouli stations at high and low solar activity. In order to test the predictive ability, the SVM was verified with different data from the training data. The results indicate that the predicted f₀F₂ has good agreement with observed data.