IETF-SSH archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Elliptic-Curve Diffie-Hellman Key Exchange draft



Douglas Stebila wrote:

Niels Möller wrote:

I recently posted a draft to add support for the use of elliptic curve
cryptography in the form of Elliptic Curve Diffie-Hellman (ECDH) key
agreement to the exchange portion of the SSH Transport Layer protocol.


What's the current patent status on that area? Is it possible to
implement any cryptography on elliptic curves without getting a patent
license from the patent holders (certicom? Others?).

Yes, it is possible and it can be easily done.

ECC as an algorithm was introduced in 1985 by Neal Koblitz and Victor Miller with no patents over the algorithm. Certicom does not hold an umbrella patent right over the algorithm. It only holds patents on some peripheral implementation or optimization techniques.

RSA has a FAQ about patents related to elliptic curve cryptography at
http://www.rsasecurity.com/rsalabs/faq/6-3-4.html.  Highlights include:

"Elliptic curve cryptosystems, as introduced in 1985 by Neal Koblitz
and Victor Miller, have no general patents, though some newer elliptic
curve algorithms and certain efficient implementation techniques may
be covered by patents. ..."
and
"... In all of these cases, it is the implementation technique that is
patented, not the prime or representation, and there are alternative,
compatible implementation techniques that are not covered by the
patents."

Elliptic curve crypto can be implemented using basic school book techniques which have no patent infringement concern. Simple standard techniques such as "window table lookup", "projective coordinate space", and "non-adjacent form wNAF" can be used for performance optimization with no patent concern. These techniques are school book techniques commonly used for RSA optimization and other multi-precision integer arithmetic computation.

Perhaps I'm missing something here?

Doesn't the above state that ECC is _not_ patent encumbered?
In other words, doesn't the above state that is possible
to create an ECC implementation that does not infringe on
any patents?

Or is it that people are concerned that the information provided
from RSA's faq may not be accurate?

I'm interested in seeing this work go forward.

Thanks,

- Joseph



Home | Main Index | Thread Index | Old Index