Research On BDS/GNSS Real-time Wide-area High-accuracy Atmospheric Delay Modeling And Augmentation Precise Point Positioning Applications

With the continuous development of multi Global Navigation Satellite System(GNSS),multi-GNSS and multi-frequency observations improve the positioning accuracy of real-time precise point positioning(PPP),which has been widely used in the field of crustal deformation,GNSS meteorology,low-orbit determination and other applications.However,compared with real time kinematic(RTK)technology,the convergence time of PPP is too long that limiting the further development of PPP.It has been proved that the high-accuracy atmospheric delay corrections can effectively shorten the PPP convergence time,it is of great importance to model the atmospheric delay in real-time precise positioning service system.At present,with the implementation of some projects,such as Crustal Movement Observation Network of China(CMONOC)and National BDS Augmentation Service System(NBASS),the distribution density of real-time reference stations is greatly improved,the station spacing is 150-300 km in China mainland.Therefore,it is significant to carry out the research of wide-area real-time high-accuracy atmospheric delay modeling with GNSS observations from real-time stations in China mainland.This paper focuses on the modeling of real-time wide-area high-accuracy atmospheric dealy and the research of BDS real-time PPP applications,and carries out studies on the model and method of extraction of atmosphere delay with multi-GNSS raw observations,method of wide-area real-time high-accuracy tropospheric delay and ionospheric delay model and augmentation PPP with the consideration of real-time atmosphere delay constraint,and develop real-time wide-area atmospheric modeling and augmentation PPP processing software based on multi-GNSS un-differenced un-combined model.With a large number of measured data,the un-differenced un-combined model and the real-time processing software is validated,the performance of BDS/GPS real-time PPP in China is evaluated and analyzed as well.The specific work and main contributions of the paper include:1.Present an un-differenced un-combined GNSS model which satisfies different signal modulation technology(CDMA and FDMA)and different frequency(single frequency,dual frequency and multi frequency),and realize the extraction of the tropospheric and ionospheric delay in a single station.By using GNSS short-and zero-baselines,we compared the performance of ionospheric delay extraction with the un-differenced and un-combined method and phase-smooth pseudorange method.The results show that performance of ionospheric delay extracted with un-differenced un-combined method is less influenced by the quality of observations at reference stations and has higher accuracy and reliability.2.Develop a method for real-time high-accuracy tropospheric delay modeling(Real-time Tropospheric Grid Point,RTGP)with considertation of hight uniform.By GNSS measurement,the accuracy of GPT2w and RTGP model with different methods in China is evaluated.It shows the accuracy of RTGP model is better than 1.5 cm over China mainland and improves 67%compared with GPT2w.3.Present a method to evaluate the effect of estimation of troposphere on PPP convergence performance by introducing tropospheric dilution of precision(TrDOP).As the satellite geometry of BDS is poor and varies slowly,it needs long time to separate the estimated parameters and results in long convergence time.In order to solve this problem,this paper proposes an ionospheric-free PPP model with consideration of priori tropospheric constraint,which effecitively improves the BDS PPP convergence performance.4.Based on the measured data of CMONOC,the real-time ionospheric delay model is established in China,the ionospheric accuracy modelled by different methods(spheric function,plannar fitting,quadratic fitting and inverse distance interpolation)is evaluated in active and quiet condition,respectively,and a real-time ionophseric model which is suitable for China is determined.The real-time ionospheric model is 2-8 TECU and 1-2 TECU in active and quiet condition,respectively.Compared with the final ionospheric product of IGS ionosphere anlaysis center CODE,the accuracy of real-time ionospheric products in China improves about 50%.Appling the real-time ionospheric model in BDS/GPS single-frequency pseudorange positioning,the positioning accuracy at 95%confidence level in China mainland is better than 1 m and 1.5 m in the horizontal and vertical direction,respectively.5.Based on the study of multi-GNSS un-differenced un-combined model,a multi-frequency multi-GNSS un-differenced un-combined augmentation PPP model with atmospheric constraint is developed.The superiority and realiblity of this augmentation PPP model is validated by using measured data form NBASS stations,and the performance of BDS-only,GPS-only and BDS/GPS single-and dual-frequency PPP in China mainland is compared and anlyazed.The experiments results show that the convergence performance of BDS dual-and single-frequency PPP is greatly improved by introducing atmospheric constraint,the positioning accuracy improvs 30%-50%and 20%-46%,respectively,the time to first fix for GPS fixed PPP solution is shortened by 14.5%.On the other hand,when BDS and GPS are combined,PPP positioning performance is greatly improved,which can realize the postioning accuracy better than 10 cm in horizontal and vertical directions in 30 min,the positioning accuracy after convergence is 2 cm and 3.6 cm for BDS/GPS dual-frequency PPP in horzointal and vertical directions,respectively,improves about 40%and 20%,compared with BDS-only and GPS-only PPP.