Research On The Key Technologies Of BDS/GPS Precise Point Positioning

Global Navigation Satellite System(GNSS)has profoundly changed the way of people's lives and greatly contributed to the social progress,playing a critical role in both civilian and military fields.Positioning missions can be accomplished with only one receiver by precise point positioning(PPP)which is flexible,low-cost and free from distance restrictions.PPP was first used for the maintenance of high-precision coordinate reference frame,and then extended to the areas of geodesy and geodynamics.The technology of PPP is of great application value which is widely used in the meteorological research,deformation monitoring,earthquake warning,LEO orbit determination and other applications.With the rapid development of GNSS,the experiment conditions for multi-constellation PPP are mature when satellite constellations and navigation signals become more abundant and the precision of final products gets higher.Benefited from more observations,stronger geometries of satellites and higher redundancies of adjustment systems,the accuracy,availability and reliability of positioning are promoted and the convergence time is reduced with multi-constellation PPP.Multi-constellation PPP has become a research hotspot in the field of GNSS.The data processing method of multi-constellation PPP is similar with that of single-system PPP to some extent,but also has its own characteristics.Due to the difference of satellite types,system reference time and hardware delays of different systems,multi-constellation PPP is affected by more systematic errors.The stability analyses and processing methods of these errors are the problems to be solved at present.Therefore,further studies of the data processing method for multi-constellation PPP are required.Focusing on the key technologies of BDS/GPS PPP,inter-system biases(ISB),hardware delays,ionospheric delays and multi-system integrations are studied in the dissertation.The main work and innovations are as follows:(1)The daily and weekly stabilities of ISB are investigated with long-term data,exploring the features of ISB obtained from different final products and different types of receivers.The results show that the daily variation of ISB is stable and the average of daily standard deviation is about 0.5ns.The weekly standard deviations of ISB in different years vary greatly.There is a system bias between ISB using the final products provided by WHU and GFZ.And the system biases of different stations are of good consistency in the same week.In addition,the weekly averages of ISB are relevant with receiver types.(2)When establishing BDS/GPS short-term ISB models,as the method of least square(LS)fails in taking the different weights of ISB data into consideration,a method based on Kalman filter is proposed in which the variance of ISB data in Kalman filter is adjusted according to the time intervals between the data and the forecasting time.The spatiotemporal correlations of ISB data are made full use of and the accuracies of ISB predictions are improved.The results show that the accuracy of ISB predictions is promoted by 29.2% with the Kalman filter model.With the priori constraints of ISB predictions generated by Kalman filter model,the averaged RMS values of static PPP solutions are promoted by 2.7% and 0.9% higher in E and U components than those with priori constraints generated by LS model,and are promoted by 10.6%,26.3% and 3.4% higher in E,N,U components than those without priori constraints,respectively.(3)In order to investigate the influence of satellite differential code bias(DCB)on uncombined PPP,the forms of parameters in uncombined PPP before and after satellite DCB correction are derived in detail,and the effects of satellite DCB on the solutions and parameters of uncombined PPP are analyzed comprehensively.The results show that with satellite DCB correction,the static and kinematic convergence time of uncombined PPP are improved by 25.8% and 21.8%,but the positioning accuracy is not remarkably promoted.In the uncombined PPP without satellite DCB correction,most of the uncorrected satellite DCB is absorbed by the ionospheric delay and phase ambiguity parameters which are biased by about 4 meters at most,however,the receiver clock offset,inter-system bias and tropospheric delay parameters are nearly not influenced.(4)Ionospheric delays in the traditional multi-constellation PPP model using raw observations may be negative values because they are biased by receiver code biases.An improved model of BDS/GPS PPP with receiver differential code bias parameters using raw observations is proposed in which receiver code biases on the first frequency of each system are constrained to zero and receiver DCB parameters are estimated.Ionospheric delays and receiver code biases are separated by the presented model.The results show that with the proposed PPP model,the average convergence time in static/kinematic mode is improved by 29.8%/21.6%,the accuracy of the estimated vertical ionospheric delays is 0.906 m,and the accuracies of the estimated GPS and BDS receiver DCB are 1.74 ns and 3.07 ns.(5)Both GRAPHIC combination model and priori ionospheric constraint model are analyzed comprehensively,and the performance of the single-frequency BDS/GPS PPP is verified.The results show that the positioning accuracy of single-frequency PPP using GRAPHIC combination model is superior to that using priori ionospheric constraint model,with static positioning accuracies better than 0.2m and kinematic positioning accuracies better than 0.5m.The positioning accuracy of GRAPHIC combination model is nearly not affected by the latitude distribution of stations,while the positioning accuracy of priori ionospheric constraint model is susceptible to the latitude distribution of stations.The positioning errors in kinematic mode of GRAPHIC combination model are discrete and consistent with normal distribution.The dynamic positioning errors of priori ionospheric constraint model are continuous,but the positioning errors in the up component are susceptible to systematic errors.(6)It is difficult to determine the relative weight ratios between the priori ionospheric corrections and the observations.A weight factor searching algorithm based on ionospheric delay constraint for multi-constellation PPP using raw observations is proposed in which weight factors are used to adjust the variances of priori ionospheric corrections.According to the principle that the sum of the quadratic forms of weighted residuals be the minimum,the optimal weight factors are searched for and the variances of priori ionospheric corrections are adjusted dynamically in order to promote the positioning results.The results of PPP in static mode show that the RMS values of positioning errors are improved from 3.96 cm to 3.40 cm and the convergence time is reduced from 76.3min to 59.9min when the searching algorithm is used,in comparison with the traditional priori ionospheric constraint.(7)The mathematical model of multi-constellation PPP is developed,and the state parameters are derived in detail.The performance of multi-constellation PPP is validated from the aspects of positioning accuracies,convergence time and different occlusion conditions.The results show that the horizontal and vertical positioning accuracies in static mode of GPS/GLONASS/BDS/Galileo PPP are 0.016 m and 0.021 m,those in kinematic mode are 0.036 m and 0.051 m,and the convergence time in static and kinematic mode are 17.7min and 18.2min.When the elevation mask is 30°-40°,the dynamic positioning accuracy of GPS/GLONASS/BDS/Galileo PPP is still better than 0.2m,while the positioning results of GPS are not continuous.The daily standard deviations of BDS and GLONASS ISB are 0.38 ns and 0.43 ns.The daily standard deviation of GLONASS inter-frequency biases is 0.06 ns.