Precise Ionospheric Modeling And Differential Code Bias Estimation From Multi-GNSS Observation

As an important part of the space atmosphere,the ionosphere has a significant impact on radio communications,aerospace and other related activities.Ionosphere is one of the main error sources in GNSS navigation and positioning,which seriously restricts the positioning accuracy of single frequency GNSS users.Meanwhile,due to dispersion characteristics of ionosphere,it is possible to use GNSS to monitor the ionosphere.Compared to traditional ionospheric monitoring technology,GNSS ionospheric monitoring technology has the advantages of high inversion accuracy,wide coverage,and continuous observation throughout the day.Thus,the GNSS-based ionospheric monitoring technology has become a hot research topic in the ionosphere field.Many scholars have done a lot of research in the field of GNSS ionospheric monitoring.The basic theory of GNSS ionospheric modeling has been established,and the daily ionospheric products has been released.However,the accuracy of GNSS ionospheric products still needs to be improved.Differential code bias(DCB)as an important bias in ionospheric observations,which can reach tens of nanoseconds.It is must be corrected in the data processing of GNSS high-precision orbit determination,precise positioning and timing.With the rapid development ofBDS and Galileo systems,accurate estimation of multi-GNSS DCB becomes increasingly important.However,currently available DCB products are still mainly limited to GPS and GLONASS systems.Thus,considering the requirements of current multi-GNSS applications,it is of great significance to carry out research on multi-GNSS ionospheric modeling and accurate DCB estimation.This thesis aims to study the theory and method of multi-GNSS precise ionospheric modeling and DCB estimation,and to provide high-precision,high-reliability GNSS global ionospheric and DCB products for global navigation and positioning users.Focusing on this goal,this paper evaluated the contribution of different ionospheric observations extraction methods and multi-GNSS fusion to ionospheric modeling,and generated near real-time ionospheric product based on IGS hourly updated observations.The DCB parameters of GPS,GLONASS,BDS and Galileo satellites are also accurately estimated,and the characteristic of GLONASS and BDS satellite DCBs are analyzed.The main work and conclusions of the thesis are as follows:(1)The main problems of GNSS ionospheric modeling and DCB estimation are summarized,and the related basic principles and methods are introduced.Different ionospheric observations extraction methods are derived in detail from the GNSS basic observational equations.The commonly used mapping functions and fitting models in GNSS ionospheric modeling are summarized.The methods of multi-GNSS DCB estimation are deduced,and the precision evaluation indexes of DCB parameters are given.(2)The contribution of multi-GNSS observation fusion and PPP fixed ionospheric observation extraction methods to ionospheric modeling was analyzed.The ionospheric modeling with GPS only,GPS+GLONASS,GPS+BDS,GPS+GLONASS+Galileo and GPS+GLONASS+BDS+Galileo observations are performed based on global IGS/MGEX observation data,and the generated ionosphere products are analyzed in detail.The results show that the differences between different ionospheric products are mainly within 1 TECU,and the contribution of multi-GNSS observations to the current global ionospheric modeling is not obvious.Compared with traditional phase smoothing pseudorange ionospheric observation extraction method,PPP fixed method has no obvious contribution to the global ionospheric modeling.However,the ionospheric product of PPP fixed method is more accurate than phase smoothing pseudorange in regional network interpolation,the accuracy improvement is about 0.5 TECU.The results of single-frequency PPP show that the positioning accuracy of PPP fixed method is higher than those of the phase smoothing pseudorange method in three coordinate components,with an improvement of about 30%,36.4%and 37.5%in east,north and up,respectively.(3)Near real time global ionospheric products is generated for real time GNSS navigation and positioning Users based on IGS hourly updated observations.The difference between generated near real time ionospheric products and CODE are smaller than 2 TECU in most of the area,and the statistical mean and STD of the difference are within 4 TECU and 6 TECU.Taking the ionospheric products of CODE as a reference,the bias of near real time ionospheric products is comparable to that of post ionosphere products.The single-frequency PPP results show that the positioning accuracy of the near real time ionospheric products is close to that of post products,which is slightly worse than that of the CODE ionospheric products.(4)The DCB parameters of GPS,GLONASS,BDS,and Galileo are accurately estimated and evaluated.The experimental results show that the estimated DCB products are in good agreement with the DCB products of DLR and IGG.The difference of DCB products between three agencies are within 0.2 ns,0.5 ns,0.2 ns and 0.2 ns for GPS,GLONASS,BDS and Galileo satellite,respectively.The STDs of estimated GPS,GLONASS,BDS,and Galileo satellites DCBs are within 0.05 ns,0.14 ns,0.11 ns and 0.15 ns,respectively,which are slightly smaller than those of DCB products of DLR and IGG(5)Taking into account the application of frequency division multiple access,a DCB estimation method considering GLONASS code inter-frequency biases is proposed,and the effect of the inter-frequency biases on DCB estimation is evaluated.According to the results,residuals of the DCB estimation without GLONASS code IFB consideration exhibit frequency-dependent systematic errors,errors that can be eliminated when the GLONASS code IFB is taken into account.The GLONASS inter-frequency differential code bias(IFDCB),defined as the difference of the sum of satellite and receiver DCB estimates without and with GLONASS code IFB consideration,is taken to represent the difference between the two DCB estimation types.The largest difference of the IFDCBs between frequency channels is greater than 7 ns.The system bias between BDS2 and BDS3 satellite DCB is analyzed based on MGEX and iGMAS networks observation data.The results show that the receiver DCB differences between BDS3 and BDS2 are close to zero for the same network,i.e.iGMAS or MGEX.However.when the iGMAS and MGEX networks are processed together,we found that the receiver DCB differences between BDS3 and BDS2 are not close to zero and present an obvious systematic bias between different networks.