[time-nuts] Timing Distribution in Mountainous Terrain
jimlux at earthlink.net
Sat Sep 11 15:29:24 UTC 2010
>> If it's far enough in the future.. Hg ion traps have a lot of
>> potential.. smaller, lower power, etc. than Cs
> Commercial availability is somewhat limited.
that's for sure.. I think all the Hg ion traps are still laboratory
curiosities.. but, 10 years from now?
A problem with Hg ion traps
> would be ROHS, unless they can be exempted or assumed to be within the
> telco exempt, which would be a legal twist on the commercialisation aspect.
And Cs or Rb don't have the same sorts of issues?
> Another aspect I have been wondering about is the trap hold-length, I
> think I recall that there was some issues relating to that...
>> I think, though, that some sort of self calibrating array using the
>> target of interest is a better scheme.. multiple receivers at each site
>> separated by some distance. Getting milliradian angular resolution is a
>> piece of cake.
> That moves the expense, and I don't think the available receivers have
> that option. They intend the spatial separation to be in kms and not m.
I was thinking about changing the problem somewhat. You'd have your
stations separated by km, but each station has several receivers and can
compare the phase of the signals. You can solve for range rate
(Doppler) and angle (delta phase), and that can go into your position
solution. (this is how we navigate spacecraft, after all, and it's also
used for a variety of target tracking systems)
Changing it from a rho-rho nav problem into a theta-theta problem
(triangulation vs trilateration). The goal is to get target position
to 10 meters, at a distance of, say, 20km, so you need angular
measurements on the order of 0.5 milliradian (0.03 degree). Offhand,
that might be easier than time to 30 ns. There are some significant
issues here.. is the pulse long enough and enough power to make the
required differential phase measurement, are there propagation issues
(refraction, diffraction, multipath) that make milliradian precision
The added hardware cost at each receiver site isn't much (compared to
site costs, etc.) especially since you probably already need at least
dual redundancy, so you can do N+1 redundancy, using 3 antenna/receivers
at each site, using 2 of them at any given time.
The wavelength at the transponder frequency is about 30cm, so with a
moderate spacing of the receive antennas (say a meter), you'll get
grating lobes and an angle ambiguity, but I think that could be resolved
with the other information available (e.g. a coarse fix)
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> and follow the instructions there.
More information about the time-nuts