Intelligent transportation systems benefit from global navigation satellite systems that became reliable and affordable means of positioning, navigation and precise timing. Their propagating signals can be obstructed by environmental conditions in near-Earth space. This paper presents a scientific approach by declaring the technology vulnerability through different solar activities such as solar flares, coronal mass ejections and geomagnetic storms varying from calm periods to frequent phenomena appearance which affect telecommunications, navigation and power systems on Earth and its vicinity.
We present a novel wavelet neural network (WNN) approximation for continuous GNSS orbit construction from precise ephemerides. The algorithms for non-parametric orbit estimation with WNN overcome limitations of small input data domain training for the GNSS satellite position computations. Simulation studies proved that WNN function overcomes traditional interpolation deficiency in better performance near the end of the interval. It is shown that WNN offers better solution in storage of data used for orbit reconstruction, but retains the computation efficiency in any function domain.