[time-nuts] Time security query

Lux, Jim (337C) james.p.lux at jpl.nasa.gov
Mon Aug 24 20:18:10 UTC 2009




On 8/24/09 1:01 PM, "Hal Murray" <hmurray at megapathdsl.net> wrote:

> 
> 
> bill at iaxs.net said:
>> These time stamps are required by government regulations in some
>> industries.
> 
> Do you have a list of the industries and/or regulations and/or what they
> require?

Cellular and other wireless use precision timing of time difference of
arrival at cell sites to meet the E-911 position reporting requirements. The
precision requirement is, I think, 10s of meters.

The power transmission folks use precision time to adjust the phase angle of
the sender and receiver of power. I don't know the exact requirements (there
is a IEEE standard 1344, though) but I would think you need better than 1
degree at 60 Hz. (few microseconds) (just found it.. 1E-7 stability for a
1pps, which must be synced within 1 us of UTC, 1 hr max outage/month)

> 
> In terms of control systems, what level of accuracy do they need?  How much
> of the problem is legal vs technical?  What happens if the GPS/clock breaks?
> Do they shut down a power plant?

Depends on the power flow. Conceivably, you'd trip off and shed yourself
from the network, which can have fairly far reaching consequences (like
blacking out a good fraction of the nation).


> 
> I don't think it would be hard for a well funded bad guy.  The normal signal
> levels are very weak so it shouldn't be hard to overpower them.  I've seen
> GPS test sets advertised.  Why would it be hard to put an antenna on one and
> overpower the real signals from blocks or miles away?

Trivial.. The signal power from orbit is very low (a few tens of watts tens
of thousands of km away). The problem is making a realistic spoofed signal.





More information about the time-nuts mailing list