Phase Ambiguity Convergence Analysis
for GNSS Precise Point Positioning (PPP)
By Simon Banville, Dr. Rock Santerre & Dr. Marc Cocard (Laval University, Quebec)
Actually, precise positioning with GNSS systems (GPS and Glonass) is mostly achieved through relative modes, which means that at least two receivers are required in order to cancel common errors at both observation sites. Recent improvements in satellite orbits and clock corrections modelling have led to the development of an alternative solution to relative positioning, namely Precise Point Positioning (PPP). This technique proposes to use a single receiver and to model the errors degrading the positioning accuracy, simplifying the logistics and costs of GNSS surveys.
The main drawbacks of this method are that carrier phase ambiguities are not integers due to a non-zero initial phase offset for each satellite-receiver pair and that the residual errors could often exceed the threshold required to usually fix the phase ambiguities to integers. Thus, traditional ambiguity resolution techniques do not apply and one must wait several minutes for the phase ambiguities to converge to obtain centimetre level accuracy. Some efforts have been made in the last few years in order to reduce this convergence time, but instantaneous centimetre accuracy is still a goal to reach.
This research focuses on three important aspects that could help in reducing the convergence time. The first one concerns the modelling of the receiver clock error, which is a nuisance parameter for positioning purposes. The second part of this research implies a study of the decorrelation techniques that can be applied to a constant non-integer ambiguity solution, which would improve the search speed and reliability of the estimation. Finally, the impacts of GNSS modernization (third frequency for GPS and Galileo) will be assessed from a geometrical point of view and by analyzing the different signal combinations.