Ionospheric TEC Modeling And Three-Dimensional Tomography Based On GNSS

The transionospheric GNSS(Global Navigation Satellite System)signals are affected by the ionospheric dispersion,and undergo path bending,delay and phase advance propagation,Different frequency signals are affected differently.The ionospheric delay suffered by GNSS signal is proportion to the TEC(Total Electron Content)along the signal path,and inversely proportional to the square of the signal frequency.In general the error of pseudorange or phase caused by the ionosphere is about ten meters,this seriously affects the positioning accuracy of GNSS.While TEC can be derived by differencing pseudoranges or phases of dual frequencies,this makes the GNSS to be an important way to monitor the ionosphere.The promotion of digitization,the advent of the 5G era,increased activity of spacecraft and other equipment in the ionosphere,higher accuracy is required for GNSS-based ionospheric monitoring.However,there are still many shortcomings in the current GNSS-based high-precision TEC estimation.One of them is that the projection function is too simple and rude,its error is widely criticized.Many scholars try to establish a new projection function,but the standard geometric projection function based on the thin layer model is still one of the widely used projection functions.On the one hand,after the ionospheric hypothesis model is complicated,the projection function becomes very complicated and even difficult to display expression,which is related to the complex structure and instability of the ionosphere.On the other hand,the refinement of the projection function may also mean that the regional difference is enlarged,and the regional difference of the global ionosphere is difficult to express with a universal and highly accurate projection function.In addition,its technical framework limits the high-precision TEC estimation based on GNSS.VTEC(Vertical Total Electron Content)model error,projection function error,DCB(Differential Code Biases)error and observation error affect each other,the total error is difficult to control.In the end,a three-dimensional ionospheric tomography model will inevitably replace it,which is more reasonable and closer to the real situation.In this context,this paper aims to study the high-precision TEC estimation method based on GNSS and the tomographic algorithm for inversion of ionospheric electron density.The main work of this paper is as follows:(1)Firstly,the principle and framework of GNSS-based high-precision TEC estimation technology are introduced.The accuracy of the four VTEC models is compared by experiments.The accuracy of the four methods under the different cut-off elevation angles and the different estimation strategies are analyzed..(2)Then,the influence of thin shell height is studied.It is further proposed to use the optimal thin shell height to improve the accuracy of TEC estimation,and the concept of optimal thin shell height is given.The estimation method is proposed.Based on the data of 11 years of solar activity period at the different latitudes IGS(International GNSS Service)sites,the optimal thin shell height was analyzed and the long-term estimation results were used to establish a thin shell height model for each site.the predicted thin shell heights are applied to DCB estimation to verify the accuracy of the optimal thin shell height model.(3)The ionospheric three-dimensional tomographic inversion study is further carried out,and the method of regional ionospheric tomography using compressed sensing technology(CSCT)was proposed for the first time.Based on the International Reference Ionosphere(IRI)model output electron density,the dictionary is established,the sparse representation method of ionospheric electron density is deeply studied.The weight matrix is used to optimize the performance of the observation matrix,and the scale factor is obtained by the NeQuick 2 model to compensate for more observations.The performance of the observation matrix is lacking.The verification experiments were carried out based on GNSS and ionospheric ionosonde in China and Europe.The results show that the CSCT method has good validity and accuracy.