Research Information System
Thesis Type: Postgraduate
Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, Turkey
Approval Date: 2024
Thesis Language: English
Student: Nurullah GÜRCAN
Supervisor: Özgür Ertuğ
Open Archive Collection: AVESIS Open Access Collection
Abstract:
There are so many applications that require small tolerance to position error. It is so critical for nearly all military and avionics systems. Enhancing the accuracy of position estimation by even a few centimeters is a significant improvement. Estimation errors occur due to a lot of error sources, like multipath effects, ionosphere, and troposphere effects, Doppler shift, clock sensitivity. Besides these sources, mathematically position calculation process after obtaining satellite positions and measuring pseudoranges also causes an error. GPS systems calculate position via the distance between receiver and satellites. An equation is obtained for each satellite the receiver monitors, and the unknowns can be solved by a set of equations. Different mathematical methods have been developed to solve the set of equations. In this study, the effects of three different position calculation algorithms used in literature are compared by means of position accuracy. Satellite positions are learned from a GPS record of a fixed GPS receiver station whose position is known, and a set of equations is formed for receiver position including satellite positions and the distances between the receiver and the satellites. This set of equation is processed with three different position calculation algorithms which are the least squares method, the weighted least squares method and the multilateration algebraic method and calculated receiver position is compared with known receiver position. In the thesis study, model parameters in exponential and sigma model for weighted least squares method are examined. It has been revealed that the weighted least squares method using the exponential model yields the lowest average position estimation error of 7.97 meters. This value indicates a 31% improvement compared to the least squares method.
Key Words : GPS, error, least squares, weighted least squares, multilateration, Bancroft, position estimation