[onerng talk] An update ....

Paul Campbell paul at taniwha.com
Tue Mar 11 10:01:01 GMT 2014


On Tue, 11 Mar 2014 20:09:40 Peter Gutmann wrote:
> Paul Campbell <paul at taniwha.com> writes:
> >I haven't gone away but I have been busy - firmware is AFAIK finished and
> >working - I'm sure bugs will surface, I've been writing technical docs -
> >they're in a partly finished state and not publicly  linked yet - please
> >
> >check them out at:
> >        http://www.moonbaseotago.com/onerng/
> 
> Just a random question here, in my copious spare time recently I've been
> doing a bit of reading about USB SDR applications, what about taking a data
> stream from a DVB-T/DAB stick and using that as an entropy source?  Since
> you're getting the raw data (before processing and error correction), which
> should be relatively unpredictable and differ over time, that would also
> seem like a good entropy source.

well yes and no - (puts on old dusty cable/satellite developer hat) noise from 
an unlocked channel would be good - bits from some active qam/transponder 
maybe not so - especially if the transport stream  is largely empty (it will 
be stuffed full of 1s and therefore lacking in entropy) - a fully active 
channel (full of compressed MPEG) might be a lot better - I'd search for 
channels that don't lock and pull the raw ADC data if you can.

> >I'm off to China at the end of the month mostly on a "learning how to
> >operate in China" trip but I'll be looking out for manufactyrers
> 
> I assume you've read the Australian Freetronics guys' writeup on this...
> 
> http://www.freetronics.com/pages/the-ultimate-electronics-hobbyists-guide-to
> -shenzhen

no, I haven't, thanks - I'll check that out

> Bunnie Huang also has various posts on his blog about dealing with Shenzhen:
> 
> http://www.bunniestudios.com/
> 
> (alongside the ever-interesting Name That Ware series).

yes (I actually won that a few months back, sort of by default) - actually I 
think Bunnie's likely to be there so I'll get a chance to say "hi" 

	Paul



More information about the Discuss mailing list