[FOSS-GPS] High accuracy positioning with low cost GPS devices:a
john at coyotebush.net
Wed May 6 19:32:48 EDT 2009
Some thoughts about time tags, satellite time and receiver time ... I can't
guarantee this is the right explanation, but it is how I think of it.
The time tag is more a label than anything else. It indicates the time the
signal ideally would have been transmitted. In practice, things don't work
perfectly, so we have to make adjustments.
The satellite clock has errors, but the errors are very predictable.
Satellite errors can be extrapolated from the ephemeris.
The receiver clock is the real problem. It is a quartz crystal which keeps
good time by human standards, but it is woefully inadequate for positioning.
Think of being a 1/4 mile off for every microsecond lost and you'll see how
important it is to have accurate clocks. A GPS receiver compensates by
tracking an extra satellite and solving for x,y,y,t instead of just x,y,z.
Including an extra satellite takes care of most of the clock error. However,
part of the clock error is drift. Drift is like having a measuring tape
which stretches; if you know how much it stretches, you can compensate. The
trick is to figure out how much the clock has drifted and to "unstretch" the
pseudo-range and phase measurements.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the FOSS-GPS