A Simple Sequential Method for Integer Ambiguity Resolution in Real-Time GNSS Positioning

Abstract

The successful resolution of integer ambiguities in Global Navigation Satellite System (GNSS) carrier-phase measurements is an essential but challenging task, especially where real-time kinematic (RTK) positioning is concerned. In this paper, a simple sequential least squares method for integer ambiguity resolution is presented in order to demonstrate how an open source toolkit - known as GPSTk - can be used by researchers as a reusable resource for developing sophisticated GNSS data processing software. GPSTk provides a set of GPS processing algorithms for static baseline computation, standard point positioning, atmospheric models, time and coordinate transformations, I/O routines, and many more. This paper describes a modification of the original solution for static GPS baselines to enable RTK positioning. A simple sequential least squares algorithm is devised and implemented to fix the integer ambiguities in kinematic mode. The LAMBDA method also employed for rapid ambiguity resolution, hence its take about 30 seconds to fix the ambiguities, and after ambiguity initialisation the resultant RTK positioning accuracy is comparable to that obtained using commercial software. The sequential method can be used not only for rover receiver initialisation, but also for network-RTK ambiguity resolution.