Analysis Of Precise Point Positioning Performance And Study On Tropospheric Models

Precise point positioning(PPP)is a high-precision positioning technology based on a single receiver with the help of products such as precise ephemeris and clock difference to achieve the positioning effect of meter to millimeter level.With the development of Global Navigation Satellite System(GNSS),the user's demand for PPP positioning is higher and higher.Due to the propagation delay of GNSS signal passing through the troposphere,how to reduce the influence of tropospheric delay on positioning accuracy has been one of the key points of PPP technology research.The zenith tropospheric delay(ZTD)is easy to model and calculate.The transformation of the actual path tropospheric delay and ZTD can be achieved by using the mapping function.Both the parameter estimation method and the model correction method can effectively correct the tropospheric delay to a certain extent,but the former will lead to the increase of the parameters to be estimated and reduce the convergence rate.While the model correction method is convenient and fast,the accuracy is low without the measured meteorological parameters.The main work and results of this paper are as follows:1.The effects of different precise ephemeris and tropospheric models on positioning accuracy are analyzed based on an open source satellite software called RTKLIB.For improving the positioning accuracy,this paper designs the calculation strategy according to the observation data quality analysis result.And then a series of PPP controlled experiments were performed.The results of the experiment show that the precise ephemeris of different analysis centers can make the PPP achieve the ideal effect,but there are some gaps in positioning accuracy,which can be selected according to the actual demand in the application.The choice of tropospheric model has a great influence on positioning accuracy.When using the Estimate ZTD model,the positioning accuracy in each direction increases by a minimum of 92.5% and a maximum of 320.80% compared with the Saastamoinen model.2.A novel fusion tropospheric model is proposed for the absence of measured meteorological parameters.The MOPS model and the GPT2 w model belong to the non-real measurement meteorological parameter model.The MOPS model can calculate the meteorological parameters at the average sea level,and the GPT2 w model can calculate the meteorological parameters at the station.Without the measured meteorological parameters,this paper proposes a fusion tropospheric model.This model uses two model methods to calculate the meteorological parameters,and then uses the empirical formula of the Saastamoinen model to solve ZTD.This model is free from the limitation of measured meteorological parameters compared with the traditional Saastamoinen model.3.Based on the above fusion tropospheric model,a novel PPP post-processing scheme is proposed.Two conventional tropospheric models and two precision tropospheric models using parameter estimation are set in the RTKLIB.The precision tropospheric model increases the parameters to be estimated for the observed equations and affects the convergence rate.Limited by measured meteorological parameters,the accuracy of traditional tropospheric models using standard atmosphere in RTKLIB is not high.This paper presents a new PPP post-processing scheme based on RTKLIB based on the above fusion tropospheric model.The experimental results show that the scheme gets rid of the limitation of measured meteorological parameters.The corresponding positioning accuracy of the PPP in the east,north and up direction is 2.25 times,1.56 times,1.43 times of the Saastamoinen model,and is2.08 times,1.39 times and 1.31 times of the MOPS model.The positioning accuracy in the up and north direction is close to the high precision parameter estimation model,which can improve the positioning accuracy of the PPP.