Research On Grid Virtual Observations Generation Based On Ground-based Augmentation System

VRS(Virtual Reference Station,VRS)technology has been widely used as a mature positioning method.However,the traditional VRS technology needs to generate a virtual observation for each user and keep it updated.This method has a large computation load on the data center and it is difficult to provide location services for a large number of users.Therefore,it is necessary to improve the method of generating virtual observations of traditional VRS.This paper proposes an improved idea to generate virtual observations of grids.Through dividing the regional grid and generating virtual observations at the grid points,the problem of small concurrency of traditional VRS users can be solved.Therefore,it has important theoretical research and engineering application value.In this paper,based on the ground-based augmentation system,an in-depth research on the method of grid virtual observations is conducted.This article focuses on the grid VRS error analysis and processing methods,interpolation model analysis and refinement,grid virtual observations generation and precision analysis as well as other related contents.The main research results and innovation points are summarized as follows:1.The grid VRS error source is introduced and the error processing method in this paper is summarized.The main analysis of double-difference ionosphere delay and double-difference tropospheric delay is influenced by factors such as height difference between satellites,satellite elevation angle,and baseline length.Taking into account the differences in the composition of GPS and BDS constellations,the magnitudes and variations of the effects of double-difference tropospheric delay and double-difference ionosphere delay for GPS and BDS are compared and analyzed.2.In-depth study of the network interpolation model,proposes to assess the robustness of the interpolated model by using error factors,and to compare and analyze the robustness and interpolation accuracy of commonly used interpolation models.There are great differences in the spatial distribution of the different interpolation models and the tolerances of the common interpolation models in the network all meet the requirements.Commonly used interpolation models have low interpolation accuracy in areas with large height differences and active ionosphere periods.3.In terms of the problem that the commonly used interpolation model has low interpolation accuracy in the midday period,the interpolation model has been optimized and improved.An ionospheric delay real-time grid interpolation method is proposed,which can improve the ionospheric delay within the Insertion accuracy.4.In terms of the problem of low interpolating precision in common high-difference areas,the interpolation model with additional height correction was tested.The results show that the accuracy of tropospheric delay interpolation for large areas with high differences is significantly improved.5.The method of generating virtual observations of grids is described and the appropriate density of grid partitions is studied.On this basis,virtual observations of grids are generated.The user-side solution is discussed.The positioning accuracy of the grid's virtual observations of the design experiment is analyzed.