[149252] in cryptography@c2.net mail archive
Re: [Cryptography] cheap sources of entropy
daemon@ATHENA.MIT.EDU (Ben Laurie)
Tue Jan 28 19:39:10 2014
X-Original-To: cryptography@metzdowd.com
In-Reply-To: <7A1D9271-C187-4253-A211-1E593538521E@callas.org>
Date: Wed, 29 Jan 2014 00:37:16 +0000
From: Ben Laurie <ben@links.org>
To: Jon Callas <jon@callas.org>
Cc: Cryptography <cryptography@metzdowd.com>,
Christian Huitema <huitema@huitema.net>,
Bill Frantz <frantz@pwpconsult.com>
Errors-To: cryptography-bounces+crypto.discuss=bloom-picayune.mit.edu@metzdowd.com
On 27 January 2014 21:14, Jon Callas <jon@callas.org> wrote:
> Assume a frame F_i taken from the camera. If the attacker cannot guess the *exact* contents of that frame, even with filters, in-camera JPEG, etc. with an advantage over guessing flips of a fair coin, then it's got one bit of entropy.
>
> So take 512 frames and hash them. Poof you're done.
>
> Extrapolate from there to the case of a frame having N bits of entropy.
Where N, it should be noted, can be less than one. Which makes
extrapolation hard.
_______________________________________________
The cryptography mailing list
cryptography@metzdowd.com
http://www.metzdowd.com/mailman/listinfo/cryptography
