[Cryptography] "On-chip random key generation done using carbon nanotubes"
Jerry Leichter
leichter at lrw.com
Tue Feb 23 18:14:04 EST 2016
> Interesting... It is an on silicon one time pad that is not shared.
> It should be nonvolatile and if permitted by hardware it can be mapped.
> If it can be mapped a second copy of a one time pad can be made to phone home
> with. A hard fuse link set could make it near impossible to map a second time
> thus making it an interesting mobile resource.
Not sure what you are talking about here. This is not a one-time pad; it's a unique, random, unpredictable key, or ID, or whatever you want to make of it. But it certainly wouldn't be used as a one-time pad. The test chip they built had 64 bits on it - not really even enough for a decent key today. While I can easily see them scaling up to, say, 256 bits or even a few Kb, they don't seem to be talking about millions or billions of bits. And even if they were - we have plenty of ways of *generating* secure one-time pads today. This would be a bizarre and complex technology to solve the easy part of the problem - the hard part, as always, being key distribution.
> Large is very possible.
> Fragile? If too fragile data gets lost when it should not be lost.
You might want to read the article on the sense in which the data is "fragile".
-- Jerry
More information about the cryptography
mailing list