Font Size: a A A

Research On The Precise Correction Model Of Regional Tropospheric Delay

Posted on:2021-03-21Degree:MasterType:Thesis
Country:ChinaCandidate:J T LiuFull Text:PDF
GTID:2370330605459047Subject:Traffic mapping information technology
Abstract/Summary:PDF Full Text Request
Tropospheric delay refers to the signal delay caused by electromagnetic wave signal passing through neutral atmosphere,which is one of the main error sources affecting GNSS ranging.The high-precision GNSS navigation and positioning can not be separated from the hardware support of the ground receiver and the satellite terminal.At the same time,it puts forward higher requirements for the rigor of the data post-processing algorithm and the reliability of the correction model involved.After more than 50 years of development,the tropospheric delay correction model has changed from an empirical model relying on measured meteorological parameters to a numerical model without meteorological parameters,and from a global model to a regional model.At present,many regional tropospheric delay models are not real-time because of their low time resolution.In addition,the fitting function of the model is relatively simple,so it is difficult to accurately describe the temporal and spatial variation of tropospheric delay.In this paper,California is taken as the experimental area to study the variation law of the tropospheric delay sequence at different stations of CORS network.Based on this,a new regional tropospheric delay time-space modeling method is proposed to establish the tropospheric delay prediction model in California.The model is tested with the measured data and analyzed in the application of single point positioning.The work and achievements are as follows:1.The theory of tropospheric delay is briefly introduced,and some classical tropospheric delay models and mapping functions are summarized.The operation flow of GAMIT software to calculate tropospheric delay at zenith is described.2.Based on the combination of discrete wavelet transform,AR model,SVR model and online-SVR model,a regional tropospheric delay spatiotemporal modeling method is proposed.According to the 2016-2018 tropospheric delay series of 94 stations in California calculated by GAMIT,a 1-day and 2-hour resolution tropospheric delay prediction model is established.3.The tropospheric delay models with two resolutions are tested in 24 stations respectively.On the whole,the prediction accuracy of the two models for ZHD is millimeter level and for ZTD is centimeter level.The accuracy of ZTD and ZHD predicted by two-hour resolution tropospheric delay model is higher,which is 1.1cm and 2.0mm respectively;the accuracy of prediction by one-day resolution tropospheric delay model is slightly worse,and the accuracy of ZTD and ZHD are 3.1cm and 6.1mm respectively.4.The spatial distribution characteristics of the tropospheric delay in California are analyzed by using the prediction values given by the model.It is concluded that the elevation is the main cause of the spatial distribution difference of the tropospheric delay in California.The higher the elevation is,the smaller the tropospheric delay is.In addition,from the distribution of amplitude,the inland area has obvious annual change cycle,the half year change cycle is weak;the coastal area annual change cycle and half year change cycle are not significant.5.The 1-day and 2-hour tropospheric delay models are applied to pseudo range single point positioning and precise single point positioning respectively.Compared with the case that the tropospheric delay is not corrected,the errors in X,Y and Z directions are significantly reduced,and then the point position errors are converted to N,E and U directions.It can be observed that the positioning errors in U direction are significantly improved.The experimental results show that the model has a good correction effect on the tropospheric delay of the pseudo range and carrier phase observations,and has a certain practical value.
Keywords/Search Tags:Regional Tropospheric Delay Model, Wavelet Transform, SVR Model, AR Model, Single Point Positioning
PDF Full Text Request
Related items