Return to Session B3 Next Abstract

Session B3: Future Trends in GNSS Augmentation Systems

Wide-Lane Ambiguity Support as an Enabler of Low-Cost, Dual-Band Precision GNSS Performance in Modern, Large-Scale Correction Networks
Mike Horton, HYFIX and AI Daniel Burr, Swift Navigation
Date/Time: Thursday, Sep. 22, 8:35 a.m.

In this paper we demonstrate how the implementation of support for wide-lane ambiguity resolution in low-cost, dual-band GNSS receivers can enable seamless, high-accuracy applications at a continental scale. Results are detailed using correction data from different wide area correction services in combination with different positioning engines.
Many different approaches for GNSS corrections have been devised during the past few decades, each having their own strengths and weaknesses. For mass-market applications, the ideal correction technology is one which converges quickly, is applicable across a wide area and reliably provides the desired level of accuracy. A number of commercial services available today promise centimeter-level accuracy at continental scale with convergence times measured in seconds.
Complementing the availability of commercial correction services, rapid growth can be observed in cost-effective, dual-band GNSS chipsets over the previous few years. These devices are frequently used in small, industry standard footprint modules and offer a path to achieving levels of performance previously associated with much higher cost hardware. With mass-market deployment of chipsets capable of consuming corrections, we expect to see many new verticals benefiting from higher accuracy location.
Correction data from the Skylark correction service has been optimized for the use of linear combination. The use of linear combinations such as the wide-lane combination has many advantages that are well understood by the positioning community, but Skylark is the first correction service to apply these techniques to achieve uniform performance on a continental scale, giving the user centimeter-level precision with fast convergence times.
A wide-lane combination is a linear combination of measurements from the same satellite, at the same time, on two different frequencies. The result is a signal which has a longer wavelength than the two original signals. Instead of attempting to resolve integer ambiguities using the original code/carrier observations, a GNSS receiver may instead attempt to perform ambiguity resolution on this wide-lane combination using traditional ambiguity resolution techniques. The longer wavelength of this linear combination allows more resilience against some of the typical GNSS errors. This combination then allows a positioning engine to achieve a correct ambiguity resolution solution a higher percentage of the time and using poorer input observations at the cost of a small increase in noise on the final position solution.
Testing results establish a performance baseline with RTKLIB and then demonstrate the performance improvement gained using a positioning engine which supports wide-lane ambiguity resolution. The wide-lane capable positioning engines tested were: a modified version of RTKLIB, Starling (from Swift Navigation) and HYFIX.
High-level results for Skylark corrections with RTKLIB saw an improvement in 2D Horizontal Accuracy from 0.24 m (baseline) to 0.04 m (wide-lane) and convergence times reduced from 112.44 to 2.69 seconds. Similar results were demonstrated with the Starling and HYFIX positioning engines and will be presented in the paper.
The significance of adding support for wide-lane ambiguity resolution to these positioning engines is to demonstrate how more applications can benefit from higher GNSS accuracy and precision through the utilization of a large-scale, continental-wide corrections network with low-cost, dual-band chipsets.



Return to Session B3 Next Abstract