[saag] Re: IETF Use of PUF

PUF == physically-unclonable function
    {Sometimes as simple/trivial as observing the biases in uninitialized ram.}

Hesham ElBakoury <[email protected]> wrote:
    > I appreciate your opinion on using PUF for device authentication
    > compared to other methods such as IEEE 802.1AR?

1. They are not mutually exclusive.
That is, a major use for the PUF is to act as a seed to generate a private
key.  The factory also knows the PUF, and generates the same key, and gets
the public part signed to turn it into an 802.1AR certificate.


2. PUF is fundamentally a shared secret.  So anyone who can authenticate the
   PUF can also impersonate the device.


3. There are many many different kinds of PUFs, and many people are skeptical
about a number of the methods.  Often too many NDAs around it, so they do
not, in my opinion, get investigated as much by researchers.

Please read/review: draft-irtf-t2trg-taxonomy-manufacturer-anchors-09
section 4 (4.1.1.3 specifically)


--
Michael Richardson <[email protected]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide

_______________________________________________
saag mailing list -- [email protected]
To unsubscribe send an email to [email protected]

-----BEGIN PGP SIGNATURE-----

iQFKBAEBCgA0FiEEbsyLEzg/qUTA43uogItw+93Q3WUFAmhmqYEWHG1jcitpZXRm
QHNhbmRlbG1hbi5jYQAKCRCAi3D73dDdZailB/9roYezbdq+CwUEOnbSGbbVxhz7
b07WhVIQ5qGhQ0PsSUcrQlVAp+zkpswSAHA4PF9HaDR+Lfk6s8MUHChzYm9Fmq33
JdB4qTjcPDFhs7T1KgxJfUhe+i7fzv/qBnn0HaqnnJjZKkh2+KsWBK7g2cC9LGtP
CSQsMDyVP7oLGTnro/8G/Mqn9wA7y6/AEUpcALxSiaWlb48uxaZgRvT/XVvEqalX
DARj5djM2lJxOpYNsK3WBSBVG0IohlE4TlL6MM20uylIIfQ491wIYFfG5tDBw4j+
h7nQ2BeEYwZgndA5NyO2UsI5etR3QKENq3PzcoFiJ2lgd8dz1WyVImZ33i1A
=QQyp
-----END PGP SIGNATURE-----