Security Dynamics

The MIS department where I work has started using "Secure-ID" cards
made by Security Dynamics Inc for access to their MVS systems.
After listening to a presentation by marketing droids and technical
support from Security Dynamics I had these impressions:

	The cards are programmed at the factory with a "random"
	seed.  They have an internal clock, and a lithium battery.
	They use a proprietary encryption algorithm to encrypt the
	time of day using the internal seed and display it on an
	LCD display using about 6 or 7 digits.  The display updates
	itself every 60 seconds (this frequency is adjustable when
	you order the cards)

	An authenticating host will have the cards seed, as well
	as the cards "clock offset" (the time the card was seeded,
	and the clock reset).  The user has a 4 digit PIN (personal
	identification number) known only to the host (and of course
	written on the back of the card :-).  PIN numbers must be
	unique since they are used to identify the user.  At login
	time, the user is asked to type in her PIN, as well as the
	number currently displayed on the card.  This is checked
	by the host, and if it's correct the user is authenticated.

	If used on a regular basis, the authenticating host can
	detect clock drift and will adjust it's database accordingly.

	Cards can be used across multiple "realms", but this
	nessesitates trusting the cards shared key with each host
	that wants to authenticate that card.

	The cards are timed to deactivate after some time interval
	(again, this is an option) the default lifespan is 3 years,
	they can go as high as 4 or 5, but after that, the battery
	isn't reliable.

You can probably imagine some of the problems with such a system.
First and foremost in my opinion, it uses an unknown proprietary
algorithm which is a closely guarded company secret known only to
them and anyone which a dissasembler.  Obviously such an algorithm
has never undergone any serious scrutiny.  Most respectable
researchers (outside of Ft Meade) do not need to disassemble code
to find material to write papers on.

Second, the cards are programmed at the factory, and the user has
no way of reseeding them.  The company actually touts the fact that
they have all the card info for all customers on file, and will
gladly send you encrypted tapes or floppys if you loose you database!
Of course they will only talk with one designated contact at your
site, and they will only ship materials to that person.  In all
fairness if your a big client, and you insist, they might be
compelled to tell you how to seed the cards, and give you a batch
of "raw" cards.

When I mentioned how ludicrous it was for us to trust their internal
security, they made some lame noises about employees being "bonded".
In other words, they have established plausible denyability and
are "out of the loop" should your security data be compromised.
I was a little furious.

Lastly, their expensive.  Something on the order of $60/card in
quantities of 250 to 500 for cards that last 3 years and change
every 60 seconds.  Programmable DES devices (used by DEC and others)
which employ a challenge response system are about one third as
much.

I came away from the talk with a bitter taste in my mouth.  As I
understand it (and please correct me if I'm wrong) they are, at
this point, one of the largest companies "crypto card" companies
in the world.  This is, to say the least, unsettling.

If you want more info, they're Colorado office is at:

	Security Dynamics
	5299 DTC Boulevard
	Suite 500
	Englewood, CO 80111
	Phone: +1 303 773-6519




brad

At one point, I talked to Security Dynamics, and experienced the same
reaction as the original poster.  Additionally, I found it ludicrous
that the cards were programmed to self-destruct after a period of
time, that they keep their encryption scheme so secret (though they'll
sell you source code if you fork over sufficient bucks), and that
there was no way to reprogram/reset the cards in the field.

I also couldn't stomach their "well, you just have to trust us"
response to a lot of my objections about it being a closed proprietary
system, having them know all of my key data, and not being able to
reprogram the cards.

I had much better luck with Digital Pathways, and their "SecureNet
Key" product.  This is a small (about as long and wide as the SecurID
card, but about 2.5 times as thick) hand-held authenticator that's
meant to go with Digital Pathways' "Data Defender" secure
communications front-end, though they sell the SecureNet Keys
separately, and they're not very expensive.  They use DES, are
individually field-programmable, and Digital Pathways is only too
happy to give you as much info about them as you want - for free.
With the info they gave me, it was no problem towrite code that knows
how to use the SecureNet Key.

--Pat.

Re: security dynamics 

One could perform an interesting test with one of these Security
Dynamics card.

Aim a video camera at the LCD display so that the display takes up the
full width of the image.  Hook the video signal up to a digitizer
board, and recognize the numbers that appear on the face.  Spit them
out as often as they appear.  For someone with all the equipment, this
should be a one or two evening hack.

Now, if the number changes every minute, that's a little over 10,000
samples in a week, certainly enough to determine if they are using
weak random number generation.

I'll put the data on the ftp site, should anyone actually do this.

Eric

>> Now, if the number changes every minute, that's a little over 10,000
>> samples in a week, certainly enough to determine if they are using
>> weak random number generation.

1) not true.  I read an article about a pseudorandom number generator
which appeared random to every test they used on it.  Then they went
and did a monte carlo simulation of something based on that prng.
Guess what?  It wasn't quite random enough.  Lesson: it can be *very*
hard to determine randomness.

2) The sequence is not random.  It is cryptographically pseudorandom.
This is very different.

3) A friend who has a significant math background in crypto stuff has
seen the Security Dynamics algorithms (under non-disclosure), and says
that they're credible.  That vouches for their theory.  That they
insist on programming the cards and keeping the keys themselves, and
that they do not allow you to program the cards yourself, is a major
problem, no matter how good their math is.

		Marc

>>> Now, if the number changes every minute, that's a little over 10,000
>>> samples in a week, certainly enough to determine if they are using
>>> weak random number generation.

>1) not true.  I read an article about a pseudorandom number generator
>which appeared random to every test they used on it.  [...] Lesson:
>it can be *very* hard to determine randomness.

The experiment I was proposing would possibly answer 'yes' to the
question "Is the number generation weak?"  It would not say how strong
it was, or even if it was strong.  it would, however, give some lower
bound on its strength or else show that it was in fact not very strong
at all.

>2) The sequence is not random.  It is cryptographically pseudorandom.
>This is very different.

Since we are talking about a device in which a sequence is duplicated
on two ends, I did not feel the need to belabor the difference between
pseudorandom and random.  The context makes it clear that this can't
be a random device based on a physically random process.

>3) A friend who has a significant math background in crypto stuff has
>seen the Security Dynamics algorithms (under non-disclosure), and says
>that they're credible.  

That bit of information may mean that a 10^4 sample test is not worth
doing.

>That vouches for their theory.  

That changes our trust from no trust at all into trust in your friend's
ability and your assessment of it. :-)

>That they
>insist on programming the cards and keeping the keys themselves, and
>that they do not allow you to program the cards yourself, is a major
>problem, no matter how good their math is.

Granted.  Their keeping the keys is worth, say, using a linear
congruential generator (or worse) in terms of overall security.

I was merely curious as to whether they were fools on all fronts, as
opposed just to the secrecy front.

Eric