Re: IP address <--> Global Positioning System (GPS)

From: Will Packard (will@nortelnetworks.com)
Date: 05/28/02 

From: Will Packard <will@nortelnetworks.com>
Date: Mon, 27 May 2002 19:11:54 -0400

Alan wrote:
>
<snip>
> The paper talks of pinning down a location to within "centimeters if
> required". This just identifies a point in space and time. The
> server recognizes the observed drift in real time at its own location
> (and perhaps other locations), and therefore knows approximately what
> to expect the drift to be at that moment at the remote location,
> within certain tolerances. The raw signals drift enough,
> unpredictably enough (due to orbital drift, not just Pentagon
> manipulation) to make it (allegedly) infeasible for the client to
> simulate the raw signal in real time, within the required tolerances.
> (Remember the location authentication is only considered valid for
> five milliseconds).
<snip>

Satellite orbital drift is somewhat random. OTOH, the resulting changes
in delay and doppler shift are *very* predictable when the latest
positioning information is determined, and can be modeled with very modest
hardware.

I don't know if ionospheric effects come into play or not, but they may be
much more difficult to model. I wouldn't bet my reputation on it, though.

Will

Relevant Pages