NovAtel's Annual Journal of GNSS Technology Solutions and Innovation

Issue link:

Contents of this Issue


Page 20 of 51

novatel correct 2014 For more Solutions visit velocity 21 stantly determined. Of course the receiver an- tenna co-ordinates are typically not known and the PPP filter must estimate them at the same time as it is estimating the ambiguities. This cre- ates a circular dependency: the ambiguities are only improved by improved coordinates, but the coordinates are only improved by improved am- biguities. This coupling between the ambiguities and the coordinates takes time to resolve. This time is the convergence period. One avenue for improving convergence is to improve the geometry of the solution, and the best way to do this is by adding additional sat- ellites. Figure 2 illustrates how this improves the solution: the uncertainty in the ranges from the satellite leads to uncertainty in the position. This region of uncertainty is reduced as ad- ditional ranges are introduced. The TerraStar PPP feed includes corrections for both GPS and GLONASS, maximizing satellite availability and giving the best solution geometry. Convergence Performance Individual convergences, even back-to-back, can have significantly different behavior. Because of the variation in individual con- vergences, convergence performance is nor- mally evaluated using some type of aggregate statistic. Figure 4, for instance, shows the 95th and 68th horizontal convergence percentiles for the same data used in Figure 3. Corresponding convergence times for several horizontal error levels are given in Table 1. The data for Figure 4 and Table 1 was collected under near-ideal conditions at NovAtel's Calgary headquarters. Actual customer performance will differ, depend- ing on local observing conditions and nat- ural variations in correction feed quality. Converged Solution Accuracy Once converged, variations in PPP po- sition accuracy are largely due to the correction feed quality. The TerraStar correc- tion feed has outstanding quality, yielding PPP positions with excellent stability. Figure 5, for instance, shows the post-convergence variation in horizontal errors for a one week period. The error never exceeds 20 cm. The corresponding cumulative sum percentages in Figure 6 show that the 95% horizontal error is less than 6 cm. Improving Re-Convergence When the signals to the GNSS satellites are in- terrupted, the unknown starting point of the carrier phase range measurement changes and the corresponding ambiguity must be reset. If this only occurs on a few satellites, then the re- maining, uninterrupted signals will be enough to maintain a position such that when tracking is re-established to the lost satellites the ambi- guities will converge quickly. Unfortunately, if tracking is lost to too many satellites, the PPP filter will lose its position, and both it and the ambiguities will have to converge again. Done naively, this would be a Horizontal Error Time 50 cm 3.6 min 20 cm 13.2 min 15 cm 19.8 min 10 cm 35.6 min taBle 1: 68th percentile horizontal convergence times TerraStar Reference Network. BRAZIL UNITED STATES CANADA COLOMBIA MEXICO VENEZUELA BOLIVIA PERU CHILE ARGENTINA CONGO KENYA ANGOLA NAMIBIA TANZANIA BOTSWANA MADAGASCAR SOUTH AFRICA ALGERIA LIBYA MAURITANIA MALI NIGER CHAD EGYPT SUDAN ETHIOPIA SAUDI ARABIA ICELAND NORWAY SWEDEN FINLAND UNITED KINGDOM FRANCE GERMANY POLAND UKRAINE SPAIN KAZAKHSTAN RUSSIA MONGOLIA INDIA AFGHANISTAN PAKISTAN NEW ZEALAND THAILAND CHINA AUSTRALIA PAPUA NEW GUINEA TURKEY IRAQ KOREA JAPAN ITALY INDIAN OCEAN SOUTH ATLANTIC OCEAN NORTH ATLANTIC OCEAN PACIFIC OCEAN PACIFIC OCEAN SOUTH PACIFIC OCEAN 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 2 1 1 1 1 1 2 2 2 2 2 2 2 1 2 2 2 2 1 1 2 GPS GPS + GLONASS " " PPP is at the apex of GNSS error mitigation

Articles in this issue

Links on this page

view archives of Velocity - 2014