Jay originally mentioned/implied voice having doppler was the concern, not
the RF carrier and subcarrier concerns.  He clarified later that it might be
a channel spacing or similar concern: this is much more likely.

You are partly right: the nominal RF bandwidth per se isn't a doppler
concern, but the phase-locked loops on carriers, carrier spacings, and/or
subcarriers in some similar RF systems are very narrow band control loops so
that just a few Hertz of doppler *might be* a loop bandwidth (ie, correction
rate) concern... GSM might use those control structures, and those would
have doppler specs applying to bit or packet error rates.  I quit a GSM
service for unrelated reasons, but my Sprint PCS (different scheme) has only
a few locations here with physical cell boundary sensitivities, and no form
of user-discernable doppler up through commercial aircraft speeds.  The
physics of whether the voice level is affected by doppler or whether the
modulation schemes might be is interesting.  Some PhDs have devoted their
professional lives to this kind of stuff and written books on the subtleties
of channels and these "modems".  If any doppler sensitivity is there in GSM,
it might have been an interesting technology trade-off decision made long
ago for the standards and economy.  Doppler sensitivity is not a "necessary"
condition in digitized voice schemes or we'd actually hear some effect in
many kinds of military aircraft and NASA spacecraft scenarios.


Chuck Cole

> -----Original Message-----
> From: tcwug-list-admin at tcwug.org [mailto:tcwug-list-admin at tcwug.org]On
> Behalf Of Daniel Taylor
> On Tue, 23 Jul 2002, Chuck Cole wrote:
> > Existence of a voice level doppler shift in a supposedly
> digital scheme like
> > GSM is hard to accept...
> >
> I think he is referring to doppler on the RF signal possibly causing a
> loss of the signal. However, given sources moving at
> .00000002C I don't
> think we are looking at a doppler shift that exceeds the RF band width
> of the connection.
> --
> Daniel Taylor
> dante at plethora.net
> _______________________________________________