From: Carl Ellison <cme@TIS.COM>
Date: Wed, 23 Aug 95 15:31:19 PDT
To: Andrew.Spring@ping.be
Subject: Re: Random Hiss from Mac mike
In-Reply-To: <199508232105.OAA27566@comsec.com>
Message-ID: <9508232228.AA17098@tis.com>
MIME-Version: 1.0
Content-Type: text/plain


>Date: Wed, 23 Aug 1995 19:09:35 +0100
>From: Andrew.Spring@ping.be (Andrew Spring)



>I've been looking at using the Mac's Sound Input Manager for hardware RNG.

[...]

>I suspect there's much less entropy in that signal than even this estimate,

[...]

>Does anybody have any experience/advice in this area?

I would try injecting noise and seeing how well you can control the output
signal.  I would also do a Fourier transform of the output and look for
cyclic behavior -- then see how the frequency spectrum can be modified by
turning machinery on and off, turning radios on and off, ....

You can also try various compression algorithms to see how much entropy
they claim to see.

When that's done, you can then use some of my ranno conditioning code,
free on the net (although written for UNIX, for now):

	http://www.clark.net/pub/cme/html/ranno.html

I especially like running a pseudo-random stream through ranM and then a
hash, to obliterate any patterns which might sneak past your analysis.

 - Carl

+--------------------------------------------------------------------------+
|Carl M. Ellison    cme@acm.org    http://www.clark.net/pub/cme/home.html  |
|PGP: E0414C79B5AF36750217BC1A57386478 & 61E2DE7FCB9D7984E9C8048BA63221A2  |
|  ``Officer, officer, arrest that man!  He's whistling a dirty song.''    |
+----------------------------------------------------------- Jean Ellison -+