[time-nuts] Next Generation Time/Frequency Standards May Require Provisions Preventing Vertical Displacement

Hal Murray hmurray at megapathdsl.net
Wed Sep 29 18:19:05 UTC 2010


> More importantly, does this impose an upper limit on data transport speed
> over  networks, in particular wireless networks?  If and when one produces
> the network  technology that would demand the accuracy and precision of
> these new  standards, if one object is moving relative to the other, there
> could be loss of  data as the moving clock goes out of sync with the
> stationary one.  We know this  was possible just from special relativity,
> but motion at "normal" speeds does  not contribute appreciably at the
> currently achievable accuracies and  precisions.  However, with the
> next generation, driving in a car or certainly  flying in a plane will
> limit bandwidth.  And, of course there's the gravitation  effect to contend
> with in the future as well, which could also limit bandwidth.

That's not a serious problem.  The receiver doesn't need a free running local 
clock that is exactly matched to the transmitter clock.  It can use a PLL to 
derive a local clock from the received signal.  Most (all?) Ethernet 
receivers and modems already work this way.

Think of the Doppler shift from a train horn.  If the train was sending morse 
code, you could still decode it if the train was moving toward or away from 
you.  It's just different from what you would hear if the train wasn't moving.


> As I am thinking about this, does this impose a limit on GPS accuracy and
> precision based on the next gen technology? 

Relativity is already significant for GPS.  The transmitters in the 
satellites arepre-corrected so it comes out right after the signal gets to a 
receiver on the surface of the Earth.   See the recent discussion (a day or 
two ago) on the exact value of the frequency sent by the GPS satellites.  
Once you figure out what you want, you can probably figure out what 
corrections you need to get it.

If you are comparing clocks at the level of national labs, you already have 
to correct for elevation of the receiver.  Boulder (NIST) is at 5000 feet.  
That's the same elevation used by Poject GREAT.
  http://www.leapsecond.com/great2005/

Daniel Kleppner's, "Time Too Good to Be True" in Physics Today, March 2006, 
pointed out that atomic clocks will soon define what sea level means.


-- 
These are my opinions, not necessarily my employer's.  I hate spam.






More information about the time-nuts mailing list