[8440] in cryptography@c2.net mail archive
NONSTOP Crypto Query
daemon@ATHENA.MIT.EDU (Lenny Foner)
Sat Jan 13 14:19:39 2001
Date: Sat, 13 Jan 2001 12:33:10 -0500 (EST)
Message-Id: <200101131733.MAA21062@out-of-band.media.mit.edu>
From: Lenny Foner <foner@media.mit.edu>
To: cryptography@c2.net
In-reply-to: <3.0.6.32.20010112164016.007e8c60@pop.sprynet.com> (message from
David Honig on Fri, 12 Jan 2001 16:40:16 -0800)
Cc: foner@media.mit.edu
Date: Fri, 12 Jan 2001 16:40:16 -0800
From: David Honig <honig@sprynet.com>
At 01:37 PM 1/12/01 -0800, Ray Dillinger wrote:
>Hmmm. That sounds like a trick that could be brought up to
>date. If you get two sensitive microphones in a room, you
>[A quick contemplation of the wavelength of the sounds in question
>would put an end to that speculation I suspect. --Perry]
Maybe not, because you can use the click--- you look only at intensity
envelope, summing all frequencies essentially.
[Remember your basic science: you can't resolve something smaller than
half a wavelength. (Well, you can, with certain techniques, but things
get seriously hairy at that point, and in general the limit is half a
wavelength.) Given this, it is unlikely that you're going to figure
out whether the g or the h key was struck. If I'm wrong here, I'd like
to hear a detailed counterargument or evidence. --Perry]
Even if you can't get it -exactly- right, knowing even a small patch
of keys for each keyclick can help a lot. If you're hoping for a
password, this just cut your search space enormously (and I'll bet
that things like the space key or a shift key sound quite different
than a normal key anyway---and this doesn't even include subtle
squeaks, etc). And if you're monitoring ordinary text, what strings
are words and what aren't also apply a very strong constraint. You
may have to do a little statistics, but hey, that's not hard.
