Interpolation Estimation Method of Tropospheric Delay for Long Baseline Network RTK Based on Support Vector Machine

2018 2nd International Conference on Power and Energy Engineering

Deng, Jian^1,2 ; Xu, Miaoqiang^3 ; Yu, Xuexiang^3 ; Zhang, Aiguo^1,2

School of Computer and Information Engineering, Xia Men University of Technology, Xiamen

361024, China^1

Big Data Institute of Digital Natural Disaster Monitoring in Fujian, Xiamen University of Technology, Xia men

361024, China^2

Surveying and Mapping Institute, Anhui University of Science and Technology, Huainan

232000, China^3

关键词: Compare and analyze;    Estimation errors;    Estimation methods;    Point coordinates;    Reference stations;    Training sample;    Tropospheric delays;    Zenith tropospheric delays;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/192/1/012069/pdf
DOI  :  10.1088/1755-1315/192/1/012069

来源: IOP

【 摘 要 】

In order to solve the problem of low precision of tropospheric delay interpolation under the long-distance sparse reference station, a method of tropospheric delay interpolation estimation based on support vector machine (SVM) theory was proposed. Firstly, the troposphere was assumed to be an infinitely thin single-layer membrane, and with the known information of reference stations, the puncture point coordinates and zenith tropospheric delay of each reference station on each visible satellite in the single-layer membrane were obtained. Then, the puncture point coordinates and zenith tropospheric delay were taken as training samples to optimize the appropriate core parameters of SVM and establish the SVM model of tropospheric delay. Finally, two sets of network RTK data with different lengths were selected to compare and analyze the effect of the interpolated tropospheric delay. The results showed that under the long-distance reference station, with the SVM tropospheric delay model established in this paper, the tropospheric delay accuracy of the interpolation estimation is better than 2cm, and the estimation error is generally stable, and the interpolation effect of satellite in different systems is basically the same.