[Cryptography] Entropy of a diode

[Tom Mitchell]

On Wed, Jul 20, 2016 at 2:17 PM, Kurt Roeckx <kurt at roeckx.be> wrote:

> Hi,
>
> I'm trying to understand the entropy generated by a reversed diode
> in avalanche breakdown mode.
>

 As others noted the noise depends on the silicon process and
low noise processes are the goal of most foundries.   Thus
cheep or out of specification parts may work "better" or "worse"
depending on the circuit being used.

Adjusting the reverse bias to find the knee seems to
be important and a source of bias for a zener source.

Chaoskey uses a pair of back to back


Google search found a couple things.
Looking at the reverse space of noise reduction might prove more
educational.

http://kundoku.free.fr/N/012/Noise%20In%20Receiving%20Systems/Noise%20In%20Receiving%20Systems.pdf
http://firasaboulatif.free.fr/index_files/gaidaa%20book/Generate%20%20gaussion%20noise%20by%20zener%20diode/Random%20Electrical%20Noise.pdf

Follow the prior art in patents...
https://www.google.com/patents/US4853884

Analysis tool for a Hardware Random Number Generator
https://github.com/BartMassey/anrand

http://altusmetrum.org/ChaosKey/

https://www.phoronix.com/scan.php?page=news_item&px=ChaosKey-1.0-2016

https://www.eeweb.com/blog/extreme_circuits/simple-white-noise-generator


Two references here.
https://www.schneier.com/blog/archives/2016/04/friday_squid_bl_523.html

Tinkering:
http://holdenc.altervista.org/avalanche/
http://imotp.sourceforge.net/noise.pdf
http://imotp.sourceforge.net/noise.pdf    <--- interesting.
   From the above link : "Transistor junction noise provides a source of
true random data that can
   be sampled and stored on a computer. However, this sampled data does not
contain 1 bit of entropy
   per bit of stored data. It is therefore necessary to post process the
data samples to distil the existing
   entropy into a smaller number of high entropy bits."





-- 
  T o m    M i t c h e l l
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20160721/d0941cc7/attachment.html>