Nonlinear prediction of the ionospheric parameter f(o)F(2) on hourly, daily, and monthly timescales

An application of nonlinear radial basis function (RBF) neural networks (NN s) to improve the accuracy of the prediction of ionospheric parameters is p resented. Principal component analysis is also adopted for the purposes of noise and dimension reduction, Hourly, daily, and monthly predictive models have been created for the Slough, England, United Kingdom, f(0)F(2) time s eries. The quality of the model predictions is evaluated by comparison with corresponding predictions from reference persistence or recurrence models, Each RBF NN offers a significant improvement over the performance of the c orresponding reference model. The noonday model gives a performance improve ment of similar to 6% over the baseline persistence model, For a 1 day ahea d prediction. For a I hour ahead prediction the hourly model offers an impr ovement of similar to 45% over the baseline 24 hour recurrence model. Final ly, the monthly median model gives a performance improvement of similar to 40% over the baseline persistence model, for a 1 month ahead prediction.