James A. Donald (jamesd@echeque.com)
Sat, 23 Jan 1999 20:55:34 -0800
--
At 09:44 PM 1/23/99 -0600, Mike Rosing wrote:
> Given that it passes DIEHARD randomness tests, I'll make
> the claim that there is no compressor which can find an
> algorithm which could reduce the number of equivalent bits
> from a hardware random bit generator.
Here is a counter example:
My NSA approved hardware random number generator generates
the sequence 1, 2 3, 4, 5 modulo 2^160. (it is not very
random)
These numbers are fed into SHA.
DIEHARD then attempts to find non randomness in the output
from SHA. It will fail.
However I can easily construct a special case compressor that
will compress the output from the generator to a single large
digit.
(And so, of course, can the NSA.)
Of course it would be easy for humans, though not DIEHARD, to
detect that there was something funny about that generator.
So here is a better NSA approved generator.
It generates fifty truly random bits from a true noise
source, one that draws upon the fundamental randomness of the
universe. It then encrypts them using a secret key known
only to me, the NSA, and my compressor. The other 110 bits
are simply the original bits encrypted and repeated.
Knowing the secret key, I can easily construct a compressor
that will compress each 160 bit group into a fifty bit group,
but since DIEHARD does not know the key, it cannot.
--digsig
James A. Donald
6YeGpsZR+nOTh/cGwvITnSR3TdzclVpR0+pr3YYQdkG
UAj8cFKK0XWbj/Hlj266Av6fkyy+2WaDgvGnck3N
4556dosdEpsirQgnGkt/HL0dk/YeDamYvF4dCtyI9
-----------------------------------------------------
We have the right to defend ourselves and our property, because
of the kind of animals that we are. True law derives from this
right, not from the arbitrary power of the omnipotent state.
http://www.jim.com/jamesd/ James A. Donald
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:18:05