Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Tempest attacks

12 views
Skip to first unread message

BOBES_...@delphi.com

unread,
Nov 4, 1993, 5:58:31 PM11/4/93
to
How can a tempest attack be averted? What hardware is available for
preventing this?
thanks
bob

Kurt F. Sauer

unread,
Nov 4, 1993, 10:57:54 PM11/4/93
to

Build a Faraday cage and work inside it. That should solve the problem
(unless the computers are operating when you open the door).
--
Kurt F. Sauer
Richardson, Texas

PGP key available on key servers

bd_...@titan.king.ac.uk

unread,
Nov 5, 1993, 8:15:50 AM11/5/93
to

Many harward manufacturers produce TEMPEST-shielded terminals, with an
extremely high level of effectiveness.

For details, contact either a large dealer or major manufacturers (such as ICL,
HP, etc.) direct.

TTFN
Geoff

Kevin Martinez

unread,
Nov 5, 1993, 11:38:31 AM11/5/93
to


I worked at a certain defense contactor that had every imaginable precaution
taken. We worked in GSA vaults that were 1/4" plate, 100% welded, double wall
constuction. These rooms had "airlock" style vault doors that prevented any
emissions because you could not open both at once. Even the air filters in
the A/C had EMI gasketing! I'm sure there is an applicable MIL standard.
--
The Lottery is a Tax on those who can't do Math!
- heard on UseNet
Kevin Martinez
l...@rahul.net

Donald E Eastlake III

unread,
Nov 5, 1993, 1:17:47 PM11/5/93
to
k...@netcom.com (Kurt F. Sauer) writes:
>In article <931104.64711...@delphi.com> BOBES_...@delphi.com writes:
>>How can a tempest attack be averted? What hardware is available for
>>preventing this?>>thanks>>bob
>Build a Faraday cage and work inside it. That should solve the problem
>(unless the computers are operating when you open the door).
>--
>Kurt F. Sauer

There is no general solution without knowing what the threat is. A
faraday cage will not help if you are broadcasting info acousticly
or magnetically. (tempest is normally thought of as refering only to
electromagnetic compromising emanations but the definition also includes
accoustic emanations, like if the keys on your keyboard all make a
slightly different noise when you hit them.

Donald

Bruce Schneier

unread,
Nov 11, 1993, 1:51:27 AM11/11/93
to
In article <931104.64711...@delphi.com>,

You can prevent a TEMPEST attack by putting your electronic equipment in
a Faraday cage, and filtering every line going into and our of it (power,
data, etc). This is neither cheap or easy. There are companies that sell
TEMPEST computers, TEMPEST phones, TEMPEST power equipment, TEMPEST
enclosures, etc.

Bruce

**************************************************************************
* Bruce Schneier
* Counterpane Systems For a good prime, call 391581 * 2^216193 - 1
* schn...@chinet.chi.il.us
**************************************************************************

Durval Pereira de Menezes Jr.

unread,
Nov 14, 1993, 5:35:45 PM11/14/93
to
In article <CGBF1...@chinet.chinet.com>,

Bruce Schneier <schn...@chinet.chinet.com> wrote:
>In article <931104.64711...@delphi.com>,
> <BOBES_...@delphi.com> wrote:
>>How can a tempest attack be averted? What hardware is available for
>>preventing this?
>>thanks
>>bob
>
>You can prevent a TEMPEST attack by putting your electronic equipment in
>a Faraday cage, and filtering every line going into and our of it (power,
>data, etc). This is neither cheap or easy. There are companies that sell
>TEMPEST computers, TEMPEST phones, TEMPEST power equipment, TEMPEST
>enclosures, etc.
>
>Bruce

Wouldn't it suffice to put a strong white-noise generator near the
computer? Something like these "glass lightning-balls" people use for
decorative effects?

Best regards,


--
Durval Menezes (dur...@rivendel.lin.ufrj.br)

Mark Schultz

unread,
Nov 15, 1993, 1:41:30 PM11/15/93
to
Probably not, depending on the sophistication of the TEMPEST attack. Most
systems would rely on very bandlimited detection, which is only somewhat
affected by additional noise. It would reduce the effectiveness, but
probably not render in totally ineffective. Bottom line, shield the
equipment. BTW, noise at this level is likely to adversely affect your own
equipment.


Mark Schultz [msch...@watson.ibm.com] or [m...@henry.ece.cmu.edu]
Guitar, Sound Engineering, Ultimate, SFB, Corvettes, Magnetics, and The Lord.
NOT in that order. This .SIG subject to change at any time.

R Agent

unread,
Nov 16, 1993, 6:18:50 PM11/16/93
to
In article <CGJqL...@hawnews.watson.ibm.com> msch...@watson.ibm.com

(Mark Schultz) writes:
>In <O36I...@math.fu-berlin.de>, dur...@rivendel.lin.ufrj.br (Durval
>Pereira de Menezes Jr.) writes:
[...]

>>Wouldn't it suffice to put a strong white-noise generator near the
>>computer? Something like these "glass lightning-balls" people use for
>>decorative effects?
[...]

>Probably not, depending on the sophistication of the TEMPEST attack. Most
>systems would rely on very bandlimited detection, which is only somewhat
>affected by additional noise. It would reduce the effectiveness, but
>probably not render in totally ineffective. Bottom line, shield the
>equipment. BTW, noise at this level is likely to adversely affect your own
>equipment.

Actually, white noise RF generators are apparently among the tools one
can use to defeat TEMPEST attacks. I don't know any of the tech
details behind it, but this was told to me by someone who does
freelance TEMPEST securing for corporations. He gave explained some
of the details of this to me (and others who attended HoHoCon 92).

Don't ask me for more details as I don't have them. But it does
apparently work, or is at least partially effective if done right.

RA

ro...@ccs.neu.edu (Rogue Agent/SoD!/TOS/KoX) - pgp key on request
-----------------------------------------------------------------
The NSA is now funding research not only in cryptography, but in all areas
of advanced mathematics. If you'd like a circular describing these new
research opportunities, just pick up your phone, call your mother, and
ask for one.

Ross Anderson

unread,
Nov 17, 1993, 5:54:43 AM11/17/93
to
In <O36I...@math.fu-berlin.de>, dur...@rivendel.lin.ufrj.br (Durval Pereira de
Menezes Jr.) writes:

|> In article <CGBF1...@chinet.chinet.com>, Bruce Schneier wrote:
|> >>You can prevent a TEMPEST attack by putting your electronic equipment in
|> >>a Faraday cage, and filtering every line going into and our of it (power,
|> >>data, etc). This is neither cheap or easy. There are companies that sell
|> >>TEMPEST computers, TEMPEST phones, TEMPEST power equipment, TEMPEST
|> >>enclosures, etc.
|>

|> Wouldn't it suffice to put a strong white-noise generator near the
|> computer? Something like these "glass lightning-balls" people use for
|> decorative effects?

Noise generators are sold, and there is even a (classified) standard for them.
They cost about 500 quid from Datasafe in Cheltenham.

However, they are a bit risky, due to the near field/far field effect - you can
jam the emissions perfectly well at 10 feet but find that they are still loud
and clear at 100 feet. You can also have resonance problems, such as structural
steelwork in your building acting as a waveguide.

The only really sound defence is not to emit in the first place; use a pen
instead of a computer.

Ross

0 new messages