1994-03-18 - Re: spyproofing your house/work building

Header Data

From: tcmay@netcom.com (Timothy C. May)
To: rarachel@prism.poly.edu (Arsen Ray Arachelian)
Message Hash: 72f109bf9c3a99ff0d8627f307a038a10d443accf3bface29bf39e2e7b4c3876
Message ID: <199403182008.MAA22719@netcom9.netcom.com>
Reply To: <9403181916.AA28470@prism.poly.edu>
UTC Datetime: 1994-03-18 20:08:17 UTC
Raw Date: Fri, 18 Mar 94 12:08:17 PST

Raw message

From: tcmay@netcom.com (Timothy C. May)
Date: Fri, 18 Mar 94 12:08:17 PST
To: rarachel@prism.poly.edu (Arsen Ray Arachelian)
Subject: Re: spyproofing your house/work building
In-Reply-To: <9403181916.AA28470@prism.poly.edu>
Message-ID: <199403182008.MAA22719@netcom9.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain



I don't want to stop the fun Jason Zions and Arsen Ray A. are having, but
their ideas won't work. Not that I think IR surveillance is the
highest priority to worry about, either.

But I used to be a physicist, and some of the reasoning here is
perpetuating fallacies:

> > Or you could just put a nice incandescent light fixture between you and the
> > drapes; nice IR output from those bulbs. Halogen fixtures ought to work
> > also, but fluorescent bulbs don't produce enough IR.

Nope. Incandescent bulbs will have a broad spectrum of IR, from the
near IR to the far IR. An attacker with a IR illumination system (such
as in night-vison or FLIR systems, etc.) can illuminate in a chosen
spectral range with a power level much higher in his chosen band than
any reasonable bulb will put out. Moreover, he could of course
modulate (e.g.. chop) the illumination and lock-on only to the
modulated signal. In other words, the attacker has the choice of
wavelength and signal modulation to increase his S/N.

Still not likely, except for determined attackers and targets of high
economic or strategic value. I suspect the Waco compound was under IR
and microwave surveillance, for example. I suspect I am not, and in
any case, I don't worry about it.

(The real danger is not individual targetted surveillance, but
widespread and easy surveillance of communications and tracking of
locations, purchases, habits, etc.)

> > More importantly: attach a contact-speaker to each pane of glass, and feed
> > Top 40 radio to it. There have been reported cases of spy types bouncing
> > laser beams off windows and using the reflected beams to reproduce the
> > vibrations produced in the glass by reflected sound; in other words, the
> > window panes are large membranes which vibrate in sync with the sound that
> > hits them, so you want to override those vibrations with something else.
> 
> Won't work.  You need a random, independant source of noise.  Each
> window pane should have its own noise source attached to a speaker.
> If the speaker is tuned into a radio station, they too can tune into
> the same station, then substract the two signals giving them a fairly
> clear ear to listen in from.  A random noise source that is independant
> will do well because they can't substract it out.  If two windows in
> two different rooms also use the same random noise, they can differentiate
> between the two rooms and get the sound.

Still won't work well against determined attackers. The entropy of
English speech and the _localization_ of the speaker means several
things:

- multiple windows (or other vibrating conductors, if microwaves are
used) will have a correlated signal corresponding to the speaker,
whereas the added noise will be uncorrelated (generally...one can
imagine clever hacks to try to spoof the listeners by injecting some
correlation into the noise, but this is also detectable....you see the
point, I hope)

- speech models allow phonemes, words, etc., to be plucked out of even
noisy environments (we do it all the time....so do folks listening for
the characteristic signatures of submarines, etc.)

...

> In that case they'll probably resort to using microwaves to bounce off
> a metal item in the room which would also vibrate with any sounds in the
> room.  The only real defense against that would be strong shielding
> and a microwave detector to see if the shielding failed... the shielding
> has to be sound isolated or else the walls of the shielding could be
> used to get sound.. :-(  Pretty nasty shit, eh?

Which is why we'll eventually all plan our conspiracies with
non-speech, non-in-person methods, such as with secure telecom....

Wait! Do you think that's what Clipper and Digital Telephony are all about?

--Tim May

-- 
..........................................................................
Timothy C. May         | Crypto Anarchy: encryption, digital money,  
tcmay@netcom.com       | anonymous networks, digital pseudonyms, zero
408-688-5409           | knowledge, reputations, information markets, 
W.A.S.T.E.: Aptos, CA  | black markets, collapse of governments.
Higher Power: 2^859433 | Public Key: PGP and MailSafe available.
"National borders are just speed bumps on the information superhighway."




Thread