Knowledge base: Warsaw University of Technology

Back

Studies on the influence of stochastic properties of correction terms on the reliability of ambiguity resolution in GNSS-RTK positioning

Dominik Próchniewicz

Abstract

High level precision of the position determined on the basis of carrier phase measurements in GNSS requires the ambiguity resolution. The correct ambiguity resolution becomes more difficult in case of a real-time method based on a single observational epoch (known as instantaneous solutions). In this technique, a key role is played by the proper modeling of observation errors such as those due to the ionospheric and tropospheric refraction or the orbital errors. The use of the network of reference stations allows to model such errors and makes possible the generation of ionospheric and geometric network corrections, the so-called correction terms, that are incorporated into the functional description of the positioning model. This method is known as the Network-based Real-Time Kinematic technique (Network RTK). The accuracy and reliability of the positioning model solution critically depends on the accuracy of the defined correction terms. Particularly during disturbances in the measurement wave propagation center (e.g. ionospheric disturbances or weather fronts), the modeling accuracy of ionospheric and tropospheric refraction may be significantly lowered, resulting in the occurrence of biases unaccounted in a deterministic model, the so-called residual errors. For optimal estimation of the unknowns in the positioning model it is necessary to take into account such errors in its functional or stochastic model. The Ph.D. Thesis presented here describes a method of taking accounting for the residual errors in the stochastic description of the positioning model by using the accuracy characteristics of the corrections terms. As a result, the stochastic model for instantaneous Network RTK positioning, called the Network-Based Stochastic Model, has been introduced. The research on the results of the positioning model solution which uses the proposed stochastic description allows to describe the influence of stochastic properties of correction terms on the reliability of ambiguity resolution and positioning accuracy in the Network RTK. The Thesis presents also the discussion on an optimal method of defining the reliability of rover performance. The use of stochastic properties of correction terms makes possible to construct Quality Indicators for network solution describing the availability and accuracy of the Network RTK performance. A detailed analysis of an estimation accuracy of these indicators as well as the method of presenting their spatial distribution over the network area in the form of Maps of Network Availability and Accuracy are also presented. This comprehensive study and analysis proved the dissertation thesis, stating that the incorporation of accuracy characteristic of the correction terms into the stochastic model of GNSS-RTK positioning increases the level of availability and accuracy of rover performance and allows to estimate the probability of correct ambiguity estimation for the entire area covered by the reference station network.
Record ID
Diploma type
Doctor of Philosophy
Author
Title in Polish
Badania wpływu charakterystyki dokładnościowej korekt różnicowych na poprawne wyznaczenie nieoznaczoności w pozycjonowaniu GNSS-RTK
Title in English
Studies on the influence of stochastic properties of correction terms on the reliability of ambiguity resolution in GNSS-RTK positioning
Language
(pl) Polish
Certifying Unit
Faculty of Geodesy and Cartography (FGC)
Discipline
geodesy and cartography / (technology domain) / (technological sciences)
Status
Finished
Start date
22-03-2013
Defense Date
15-05-2014
Title date
15-05-2014
Supervisor
External reviewers
Jacek Kudrys Jacek Kudrys,, External affiliation of publication: AGH University of Science and Technology
Paweł Wielgosz Paweł Wielgosz,, Undefined Affiliation
Honored
yes
Pages
194
Internal identifier
1/2014_rozprawa_doktorska
Keywords in English
Network RTK, stochastic model of GNSS observations, instantaneous ambiguity resolution, quality indicators for network solution
Abstract in English
High level precision of the position determined on the basis of carrier phase measurements in GNSS requires the ambiguity resolution. The correct ambiguity resolution becomes more difficult in case of a real-time method based on a single observational epoch (known as instantaneous solutions). In this technique, a key role is played by the proper modeling of observation errors such as those due to the ionospheric and tropospheric refraction or the orbital errors. The use of the network of reference stations allows to model such errors and makes possible the generation of ionospheric and geometric network corrections, the so-called correction terms, that are incorporated into the functional description of the positioning model. This method is known as the Network-based Real-Time Kinematic technique (Network RTK). The accuracy and reliability of the positioning model solution critically depends on the accuracy of the defined correction terms. Particularly during disturbances in the measurement wave propagation center (e.g. ionospheric disturbances or weather fronts), the modeling accuracy of ionospheric and tropospheric refraction may be significantly lowered, resulting in the occurrence of biases unaccounted in a deterministic model, the so-called residual errors. For optimal estimation of the unknowns in the positioning model it is necessary to take into account such errors in its functional or stochastic model. The Ph.D. Thesis presented here describes a method of taking accounting for the residual errors in the stochastic description of the positioning model by using the accuracy characteristics of the corrections terms. As a result, the stochastic model for instantaneous Network RTK positioning, called the Network-Based Stochastic Model, has been introduced. The research on the results of the positioning model solution which uses the proposed stochastic description allows to describe the influence of stochastic properties of correction terms on the reliability of ambiguity resolution and positioning accuracy in the Network RTK. The Thesis presents also the discussion on an optimal method of defining the reliability of rover performance. The use of stochastic properties of correction terms makes possible to construct Quality Indicators for network solution describing the availability and accuracy of the Network RTK performance. A detailed analysis of an estimation accuracy of these indicators as well as the method of presenting their spatial distribution over the network area in the form of Maps of Network Availability and Accuracy are also presented. This comprehensive study and analysis proved the dissertation thesis, stating that the incorporation of accuracy characteristic of the correction terms into the stochastic model of GNSS-RTK positioning increases the level of availability and accuracy of rover performance and allows to estimate the probability of correct ambiguity estimation for the entire area covered by the reference station network.
Project (archive)
X. Project leader: Brzeziński Aleksander, +48 22 234 76 97, start date 02-06-2014, end date 31-12-2014, 504 P 1060 0820 000, CompletedWGiKDziałalność statutowa
Thesis file
Request a WCAG compliant version
Reviews
• File: 1
recenzja dr hab. P. Wielgosz.pdf
• File: 2
Recenzja pracy Dominika Próchniewicza wykonana przez dra hab. inż. Jacka Kudrysa
Request a WCAG compliant version
Other files
• File: 1
urn:pw-repo:WUT6145bad993e949a09cfad99bf6ad7013