In the June issue’s cover story, “Interchangeability Accomplished,” is a paragraph headed, “Satellite Intersystem Biases,” which appears to assert that Galileo System Time (GST) is 3 seconds ahead of UTC.
However, in the version of the Galileo Signal In Space Interface Control Document posted at: http://ec.europa.eu/enterprise/policies/satnav/galileo/files/galileo-os-sis-icd-issue1-revision1_en.pdf, paragraph 5.1.2 appears to indicate that Galileo System Time (GST) was synchronized, at the second level, with GPS time on 22 August 1999 (that is, 13 seconds ahead of UTC). And, given that a) GST, like GPS time, does not step for announced leap seconds, and b) the IERS has, as of today, announced 3 leap seconds since 22 August 1999, such would appear to suggest that GST is presently roughly 16 seconds (vice 3 seconds) ahead of UTC.
— Stuart Eventhal
Author Frank van Diggelen replies:
Yes! You are right, the article should have said 16 seconds for Galileo, not 3. Thanks for catching that. I’ve corrected the text that appears in the online version of the article, and the accompanying figure.
The online article covering Javad Ashjaee’s input on the GLONASS situation makes a positive statement that clarifies what has been a horrible reporting job across the board by news channels.
Fox, CBS, NBC, and ABC should all be ashamed that GPS World scooped them on what appears to be a simple story.
— Mark Silver
IGage Mapping Corporation
Salt Lake City, Utah
To Consumer-Grade GNSS Chip Manufacturers
I would like you to consider including a very simple feature in your GPS functionality that will permit elevation to be identified to decimeter level in many instances. The changes needed to the chip are simply the ability to accept an accurate latitude and longitude input, and an elevation calculation function that uses input latitude and longitude.
In addition to enabling instantaneous calculation of an accurate elevation, it may be that a “residual better accuracy” will remain for some time after the calculation, and that this will permit substantially improved latitude and longitude identification at a close distance.
The geo-location scene has evolved rapidly over the past 20 years. It is now very commonplace to be able to locate the latitude and longitude of a location extremely quickly and extremely accurately. For instance, the Google Earth image from the front of my house shows the dotted dividing line in the center of the road. Measuring one of these lines in Google Earth gives a size of 3.1 meters by 20 to 30 cm wide. The lines actually measure 3.0 meters by 12 cm wide. From within Google Earth I can identify the latitude and longitude of the end point on the centre of this line to within ±10 cm with a high degree of confidence. In addition there may be some other small errors in Google’s reporting of the latitude & longitude (for example due to placement of the image or distortion of the image), but these are hopefully minimal.
Now if I place my GPS unit on the end center of this line in the road, I am provided with a result that I know is erroneous. The GPS horizontal location shown in Google Earth is very rarely within two meters of my known location. It is known that altitudinal accuracy is always some two times worse than horizontal accuracy.
If I can simply tell the GPS unit that I am at this known horizontal location, it is a relatively simple calculation to recalibrate the clock and pseudoranges to provide my elevation, which will have an accuracy of a two times the accuracy of the horizontal position. Decimeter horizontal accuracy will provide 2-decimeter altitude accuracy. This is close to 100 times better than the elevation accuracy currently available on any consumer grade stand-alone device and is also effectively instantaneous!
This functionality is simple to implement. I would hope that it could be implemented with nothing more than an upgraded ROM which includes a new API function to allow the input of “I know this is my current horizontal location” and an enhanced calculation process which uses this horizontal location to calculate altitude.
I am unsure whether a residual improvement in accuracy can be attained. Even an improved accuracy for 1 minute after the fix would be useful in many situations, and an improved accuracy for 5 to 10 minutes would be a boon.
— Glenn Thorpe