A technology study of foF2 forecasting during the ionospheric disturbance

The ionosphere of the Earth is an especially complicated and nonlinear system. Back Propagation (BP)network is applied to forecast f(o)F(2) because it is nonlinear between input and output of network. Taking into account of the different characteristics of f(o)F(2) variety with solar activities, geographical positions and seasons, we present an example of the ionospheric f(o)F(2) forecast using the ionospheric sounding data at Haikou and Changchun, China, during ionospheric disturbances in 1994 and 2001. This example indicates that the method ANN (Artificial Neural Network) for f(o)F(2) forecast is superior to IRI (International Reference Ionosphere). In addition, the residual error model is built based on the forecasting results in 2000 and 2001 at Haikou, China for different requirements of various users.