[time-nuts] U-blox teaser

[Hal Murray]

k8yumdoober at gmail.com said:
> During my Arecibo Observatory days we used NIST's TMAS service to keep our
> H-maser-based station clock synced with UTC.  Our user community (mainly VLBI
> and pulsar timing people) seemed pretty satisfied with +/- 100ns accuracy, so
> I tried to do better by keeping things well within +/- 50 ns during my reign.
>  IIRC, NIST was claiming that TMAS could produce results mostly within about
> +/- 20 ns. 

I think the VLBI guys use the time from the clock at the receiver as a 
starting point.  What they need is a constant frequency.  They can work out 
the time offset.

Here is the example I heard about.  Suppose you have N antennas.  You need to 
know thir locations very accurately.  You can work out the location of one 
antenna if you point them all at a good point source and use N-1 antennas to 
work out where that point source is in the sky and the time/angle of the 
Earth's rotation.  Then you can solve for the position of the Nth antenna that 
gives the best fit.

I assume there are iterative approaches that can be used to refine the 
positions of multiple antennas.

Position is 3D.  Time is 1D so I assume the search is reasonably quick.

There are interesting similarities/dualities between GPS and VLBI.  With GPS, 
you have N transmitters and one receiver.  With VLBI, you have one transmitter 
and N receivers.  With VLBI, you can't work out the distance to the 
transmitter.  With GPS, you can't work out the length of the antenna cable.

-- 
These are my opinions.  I hate spam.