Some comments from about a decade ago.

The way it used to work in the Army (that I
was in) within a battalion, is that there was
a little code book, with a sheet for a 6 hour
stretch. Each sheet has a simple matrix for
encoding letters, etc.  Everyone had the same
sheet, and they were created centrally and
distributed from there.  If any sheets were
lost, it was a major disaster.

All soldiers were taught to code up the messages,
it was one of the more boring lessons.  In
practice, corporals and seargeants did most
of the coding, but it was still a slow and
cumbersome process.

For most of the communications needs, soldiers
talked in the clear, using a set of code words
that never changed.  For example, Sunray is
the unit commander.  This wasn't for the purposes
of security, but for clarity.  Only reports
were encrypted.  Radios were huge, heavy, and
didn't have much facility.  They were always
giving problems, and soldiers for the most part
didn't understand their purpose (in the way
that they clearly understood what a weapon was).

I wasn't so much into professional crypto back in
those days, but thinking back, it would be a seriously
hard task to put net-quality crypto into tactical
comms.

Consider these difficulties:  it was *banned*
to use any form of comsec that wasn't centrally
approved.  No personal code words, no CB radios,
no knicknames, no nothing...  (In practice there
was some leakage, I recall on my last exercise,
logistics back to the battalion HQ in the city
was handled over a cellular phone!)

The standard radio had to be purchased from
a military supplier - like Racal - and the
procurement process was probably 4 years long
before the first units hit the troops.  During
that time there could be a revolution in the
way comsec could work, if one were to learn
anything from the lessons of SSH, etc.  Each
radio was meant to last at least 20 years...

Further, whatever was put in place had to be
handled by soldiers.  Count them as approximately
as technically adept as your grandma.  If she
can't be taught to do it on pencil and paper,
then the soldiers can't be either.

As we haven't managed to get our respective
grandma's using crypto on the net, yet, that
would suggest why the military hasn't had much
luck at the infantry level, either.

(Airforce and Navy are somewhat different of
course, as are armoured vehicles.  They have
portable infrastructure that infantry don't
have.)

Adam Shostack wrote:
> 
> On Mon, Mar 31, 2003 at 01:17:43PM -0500, Peter Wayner wrote:
> | He went on to talk about "crypto" as if it was something like fuel or
> | food. He said, "They probably loaded up 4 or 5 days of crypto at the
> | beginning, but then they had to turn it off after the supply lines
> | got muddled."

Makes sense, the troops probably carried the
code books for the next 4-5 days, but comsec
probably ruled out any more than that.  Then,
when that "ran out" the staff discovered that
the new code books couldn't be distributed to
all the soldiers.  Without all of them on the
same system, switching to clear would have
happened like an epidemic across the force.

> (Of course, if they just put the crypto on smartcards, or key fobs,
> you could likely carry a month or three worth of crypto with you, but
> then they wouldn't know what had happened to every key out there.

Exactly.  One of the things soldiers are trained
to do is, after a successful action, secure
the enemy's radios and try and recover their
codebooks or codes.  A fob or smartcard would
be just like that, a token to be captured.
Once captured, this would let one into the
net.  A big prize.

So, in practice, the commsec people would not
accept this solution.  They would know that
any pin would be listed in a plastic covered
page in the radioman's notebook.

> Clearly, its better to have unencrypted comms where you know they're
> insecure, rather than low assurance secure comms.  For some threat
> models that I disagree with, anyway.

Tactical security means where there is only a
matter of hours where the information should
be kept discrete.

-- 
iang

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]

Reply via email to