[37584] in North American Network Operators' Group
Re: DSL line stealing when there is no tone - DANGER -
daemon@ATHENA.MIT.EDU (Daniel Senie)
Wed May 16 17:19:11 2001
Message-Id: <5.1.0.14.2.20010516170556.03970ec0@mail.amaranth.net>
Date: Wed, 16 May 2001 17:14:35 -0400
To: Greg Maxwell <gmaxwell@martin.fl.us>
From: Daniel Senie <dts@senie.com>
Cc: nanog@nanog.org
In-Reply-To: <Pine.GSO.3.96.1010516164452.5312H-100000@da1server>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Errors-To: owner-nanog-outgoing@merit.edu
At 04:46 PM 5/16/01, Greg Maxwell wrote:
>On Wed, 16 May 2001, Steve Schaefer wrote:
>
> > The main reason not to stick a tone on the DSL line is that the line
> > coding (2B1Q) used by SDSL uses the baseband (low frequency part of the
> > spectrum).
> >
> > For line codings that don't use the baseband (CAP, DMT and variants like
> > G.lite), the DSLAM (telco central office DSL equipment) is always set up
> > so that the DSL can be combined with a voice circuit over the same pair,
> > so it still doesn't put a tone on the line.
>
>I predict great profits for the first person to duct tape 100 'tracer
>tone-generators' into a 23 inch rack with 48v DC power source.
Better: a chip with a recorded voice: "This pair is in use" interspersed
with a tone.
While present baseband signalling may be unable to handle such, it'd be
useful if future ones purposely avoided the zero to 5kHz spectrum to allow
for such a mechanism. Clearly the telco workers have demonstrated the
absolute necessity for such a facility.
-----------------------------------------------------------------
Daniel Senie dts@senie.com
Amaranth Networks Inc. http://www.amaranth.com
