[119117] in Cypherpunks
Re: Shielded Enclosure
daemon@ATHENA.MIT.EDU (Peter Gutmann)
Fri Oct 15 17:14:04 1999
From: pgut001@cs.auckland.ac.nz (Peter Gutmann)
To: cypherpunks@cyberpass.net
X-Charge-To: pgut001
Date: Sat, 16 Oct 1999 09:59:15 (NZDT)
Message-ID: <94002115513070@cs26.cs.auckland.ac.nz>
Reply-To: pgut001@cs.auckland.ac.nz (Peter Gutmann)
John Young <jya@pipeline.com> writes:
>We've run across a 1998 Air Force CBD solicitation below for a shielded
>enclosure which describes performance requirements. Those who may know
>if this meets TEMPEST protection requirements would perform a public service
>if confirming that is the case, perhaps by way of an anonymous message to
>punks or to me (PK appended).
[...]
>Shielded enclosure will meet or exceed an attenuation of 100Db in electric
>fields and plane waves from 14Khz to 10Ghz and will meet or exceed an
>attenuation of 60Db to magnetic fields at 14Khz rising to 100Db at 200Khz
>and remaining at 100Db or better to 10Ghz when tested in accordance with
>MIL-STD 285 and meeting all shielding requirements of the national security
>agency specifications NSA-65-6.
AFAIK this is just a general shielding requirement, NSA 65-6 is "R.F.
Shielded Enclosures for Communications Equipment: General Specification",
and originally dates from the early 1960's. If it's being used for guided
weapons testing it's more likely there's a requirement to keep things out
than in. MIL-STD 285 ("Method of attenuation measurements for enclosures,
electromagnetic shielding, for electronic test purposes or equivalent") just
tells you how to measure its effectiveness. The test procedure described in
285 involves setting up a transmitter and receiver inside a shielded room,
taking a reading in free space inside the room to use as a reference zero,
and then taking readings inside the enclosure being tested. The difference
between that and the reference zero is the shielding effectiveness. If you
want to see what else is in 285, get IEEE STD 299-1991, "Standard for
Measuring the Effectiveness of Electromagnetic Shielding Enclosures" which is
the successor to 285.
NSA 65-6 requires pretty much constant 100dB attenuation from 1kHz to 10GHz
for electric fields, and 20dB at 1kHz rising to 100dB at 200kHz, so it looks
like the requirements given just mirror NSA 65-6. In comparison, commercial
electronics gear typically has only 40-60dB of shielding.
The standards to look for if you're specifically interested in TEMPEST
shielding requirements are ones like NACSI 5004, "Tempest countermeasures for
facilities within the United States", NACSI 5005, "Tempest countermeasures
for facilities outside the United States", and the test spec NSTISSAM
TEMPEST/1/92, "Compromising emanations field test requirements -
electromagnetic". You probably aren't going to see those in an RFP :-).
Peter.
