[Cryptography] TRNG review: Arduino based TRNGs

[Bill Cox]

On Thu, Jan 14, 2016 at 2:36 PM, Ron Garret <ron at flownet.com> wrote:

>
> On Jan 13, 2016, at 5:08 PM, Jonathan Thornburg
> IMHO the right approach is to not worry about the quality of your noise
> source and just use a whitener with a large margin of safety.  Feed 1000
> raw 10-bit ADC readings into SHA512 and you’re almost certainly secure
> against any attack.  If you want to be super duper paranoid, run some basic
> sanity checks on the raw input, like make sure that the standard deviation
> of your samples is >>0.


I think you would usually be right.  An exception might be the guy who is
running in something close to a Faraday cage off a battery and good bypass
caps, with little non-thermal noise feeding the ADC.  Does anyone know what
the anti-alias filter is like on these parts?  There surely is one, and it
will limit the thermal noise.

As an example, the negative input to the op-amp on my TRNG has 24uV RMS of
thermal noise, based on the 8 MHz bandwidth and resistance.  A 10-bit
Arduino ADC probably has less than 8MHz of input bandwidth due to the
anti-aliasing filter, but I could be wrong.  If it also has 24uV RMS of
thermal noise, a 2V range, and a 10-bit resolution, it will be measuring a
24uV noise source with an ADC with 2mV resolution.  That can be a problem...

Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20160114/2e694b19/attachment.html>