[117573] in Cypherpunks
otp
daemon@ATHENA.MIT.EDU (Anonymous)
Mon Sep 6 12:39:20 1999
Date: Mon, 6 Sep 1999 18:16:21 +0200 (CEST)
Message-Id: <199909061616.SAA06037@mail.replay.com>
From: Anonymous <nobody@replay.com>
To: cypherpunks@toad.com
Reply-To: Anonymous <nobody@replay.com>
At 07:21 AM 9/6/99 -0400, John A. Limpert wrote:
>I put together a true random number generator from a Geiger counter and
>an old PC. It works well but it is a bit slow in generating random
>numbers. It's good enough for keys and OTPs for low volume traffic.
Take a $5 smoke detector. Ambient decay = 15/minute.
Smoke detector flush to window on geiger tube: 20,000 decays/minute.
>Has anyone designed a true random number generator that runs at high
>speed? Something that could fill a CD or DVD in a reasonable amount of
>time?
I have used a detuned FM radio hiss + soundcard to grab
raw data; simple parity to 'distill' (irreversibly compress);
and a variety of entropy- and structure- measuring tools to evaluate the
quality of the distilled data.
1. tune FM between stations
2. cat /dev/audio > bigfile.bin
3. while (entropy( bigfile.bin ) < desired_entropy)
4. distill < bigfile.bin > distfile.bin
5. mv distfile.bin bigfile.bin
5. burn CDROM with bigfile.bin.
brain-dead (but effective, if not the most efficient) distiller:
putchar( getchar() ^ getchar() )
.....
You could/should also use a crypto-strong transform to further
'whiten' the distilled physical entropy.
