From: Lucky Green <shamrock@netcom.com>
Date: Sat, 7 Dec 1996 15:12:17 -0800 (PST)
To: cypherpunks@toad.com
Subject: Re: Laptops and TEMPEST
Message-ID: <3.0.32.19961207151252.0068e328@netcom14.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain


[Sorry, I don't have the attribution for the fist quote. The remainder of
the post was authored by a friend of mine who retired from decades in
military SIGINT. He allowed me to forward it to the list.]

>>> don't emit enough radiation for Bad Guys to read it.  CRTs are well known
>>> as emitters of easily decoded signal, but people have occasionally
>suggested
>>> on this list that laptop LCD screens are much quieter.  I now have a data
>>> point on this one, and basically, it ain't so.


    Look - if it uses electricity; if there is an oscillator
    anywhere innit; if there is a ground loop in the circuit design,
    if there are make/break contacts anywhere - It will 
    radiate, and the amount it radiates is directly proportional
    to the basic power source ...  And circuit board traces are
    getting so damned close that engineers I know/knew were
    worried about friggin' arc-over at 3 volts.  By now, may be
    even closer and voltage worries lower...

    So we have a problem. The only computer I know of that
    is (at this date) leak-proof is the biological one
    'twixt one's ears.  And soon, maybe not even that.  Those
    who laugh at the paranoiacs who wear aluminum helmets and
    wear shoes with a static strap to the sidewalk may
    be laughing out the other side of their moufs too soon...

    As for any government directives requiring companies to make 
    computers leak, I know of none, but CAVEAT: I've been retired
    from source info since '91, and most of my friends/fellow engrs
    who occasionally got together for a few brewskis and BS sessions
    over in Mt. View have either gone to better jobs (more $$$),
    transferred back east, or inconsiderately died.  So, who
    knows? 

    And of course, the ability to detect this RF/RFI/EMI leakage    
    from your information processor is similarly dependent
    on the sensitivity of the equipment you're using for 
    detection.  I've seen absolute magic performed using 
    a Wullenweber antenna, and that is NOT state-of-the-art
    equipment any more (ca. 1965-70), even tho the DoD keeps throwing 
    money at it in upgrades (affectionately called 'the 
    elephant's cage' by those who worked with it).

    If you really need to keep your information processing
    "private", then you can either isolate yourself inside
    a double-shielded room of solid copper; power everything
    with batteries; have no wires leading out of that room;
    make damned sure the door has the nice secure wiping strips
    to complete the shield when you close it -- or move your
    information processor into the middle of a whole bunch of
    the same or worse RF/RFI/EMI emitters, and just _maybe_
    your data will get lost or become inaccessible because
    of the overload of the detection equipment by much larger
    interference fields.

    Or use paper, pencil, and one-time pads and burn everything
    that's done "in the clear" and really scrunch the ashes into
    dust.

[Again, I am not the author of the above post.]


-- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred
   Make your mark in the history of mathematics. Use the spare cycles of
   your PC/PPC/UNIX box to help find a new prime.
   http://ourworld.compuserve.com/homepages/justforfun/prime.htm