[118742] in Cypherpunks
Re: Unplugged! The biggest hack in history
daemon@ATHENA.MIT.EDU (John A. Limpert)
Thu Oct 7 02:17:21 1999
Message-Id: <199910070600.CAA09136@mail1.radix.net>
Date: Thu, 07 Oct 1999 02:00:34 -0400
From: "John A. Limpert" <johnl@radix.net>
To: Sean Roach <roach_s@mail.intplsrv.net>, cypherpunks@algebra.com
Mime-version: 1.0
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit
Reply-To: "John A. Limpert" <johnl@radix.net>
>From: Sean Roach <roach_s@mail.intplsrv.net>
>
> Flaws as I see them.
>
> I'm assuming, probably erroniously, that a given packet can be
> checked for errors, up to a certain time, where that time is before
> the packet has been completely transmitted.
You could compute the probability of having received an error-free packet
when all but the last N (N < 16) bits have been transmitted. This would
involve checking the CRC result for the 2**N possible values of the
remaining N bits. There are probably better ways of doing it than brute
force.
> Toying with the connection may tip off the bugged that someone is
> evesdropping.
A simpler method would be to install a bidirectional delay line. This would
allow you to preview the data at the input to the delay line, evaluate it
and insert hits at the output of the delay line. A TDR (time domain
reflectometer) would see a unusually long line between the subscriber and
the central office.
> Paranoid thought, and without any technical basis whatsoever. Could
> this be the reason, or part of the reason, that 56kbps modems are
> limited to a maximum speed that is less than the hardwares
> capabilities? I had assumed that perhaps the government has merely
> disallowed the use of 2.6MHz.
My understanding is that the FCC limit is concerned with crosstalk caused by
excessive modem transmit power. Years ago, when modems where bigger than
PCs, I used to adjust modem output levels to -10 dBm as part of the
installation and alignment procedure.
