Research on methods of regional ionospheric delay correction based on neural network technology

摘要

A new model for regional ionospheric delay correction has been developed, tested and compared with a two-dimensional polynomial ionosphere delay correction model (called 2-DPM). First, available ionospheric data have been used in the modelling of the 2-DPM. Then, artificial neural network technology is used to compensate for the model deviation △y of the 2-DPM to obtain a new ionospheric delay correction model (called the Fusion model). Finally, an example is implemented to demonstrate the performance and efficiency of the Fusion model. According to the statistical analysis of average root mean squared errors, the Fusion model offers an improvement of 26.4% over the 2-DPM. The results show that the proposed model can give better accuracy than the 2-DPM.