Researches On Ionosphere Modeling Of Global Navigation Satellite System

Ionospheric delay error is one of the main sources of error that affecting the accuracy of satellite navigation and positioning. The complex variation and physical structure of ionosphere lead it difficult to eliminate ionospheric delay errors and model the ionosphere with highly precision. The total electron content (TEC) is proportional to the ionospheric delay as a typical parameter to characterize the ionosphere. It has became the key and difficult points to accurately describe the spatial and temporal distribution and variation of TEC, model and forecast the TEC with high accuracy so as to improve the accuracy of navigation and positioning.This paper focused on the researches of the long-term forecasting methods of TEC and optimization algorithm of ionospheric delay model. In this thesis the characteristics of temporal and spatial distribution of the global TEC and impact that produced by solar activity was analyzed, the factors that affecting the accuracy of the spherical harmonic model was discussed, including the model order and the data sampling interval. On this basis, as the autonomous navigation having a demand for long-term forecast of TEC, the use of expansion harmonic function model and piecewise method to forecast long-term TEC was explored combined with the various periodic changes of TEC, the space-time applicability of the forecasting model was verified at high and low solar activity, the effect of different length of the history of VTEC sequence on the forecasting model was studied. Aiming at the ionospheric delay correction of single-frequency receiver, this thesis proposed an optimized low-order Spherical Harmonic Function Model. According the characteristics that selecting coefficient can be encoded, method and steps of using genetic algorithms to optimize coefficients were presented and the accuracy analysis of the optimized model was made.The results showed that:using the expansion harmonic function model and piecewise method to forecast long-term TEC can forecast the variation trend of TEC for more than 90 days. Furthermore, the longer the historical TEC sequence which was used to model, the higher accuracy of the forecast accuracy and can keep a long time without divergence. The accuracy of the optimized model was improved. It has an improvement about 1-2TECu than the low-order Spherical Harmonic Function Model with the same number of coefficients and can better reflects the daily change of VTEC and the characteristics that varies with latitude.