Ionosphere tomography using wavelet neural network and particle swarm optimization training algorithm in Iranian case study

Computerized tomography provides valuable information for imaging the ionospheric electron density distribution. We use a wavelet neural network with a particle swarm optimization training algorithm to solve pixel-based ionospheric tomography. This new method is called ionospheric tomography based on the neural network (ITNN). In this method, vertical and horizontal objective functions are minimized. Due to a poor vertical resolution of ionospheric tomography, empirical orthogonal functions are used as vertical objective function. For numerical experimentation, observations collected at 38 GPS stations on 2 days in 2007 (April 3 and July 13) from the Iranian permanent GPS network (IPGN) are used. Ionosonde observations (φ = 35.7382°, λ = 51.3851°) are used for validating the reliability of the proposed method. The modeling region is between 24°E to 40°E and 44°N to 64°N. The results of the ITNN method have been compared to those of the international reference ionosphere model 2012 (IRI-2012) and the spherical cap harmonics (SCHs) method as a local model. The minimum relative error for ITNN is 1.41% and the maximum relative error is 24.03%. Also, the root-mean-square error of 0.1932 × 10 11 (el/m 3 ) has been computed for ITNN, which is less than the RMSE of the IRI-2012 and SCHs method. The comparison of ITNN results with IRI-2012 and SCHs method shows that the proposed approach is superior to those of the traditional methods.