Re: [Mobopts] New Draft on Proxy SEND

Hi Jim,

I don't think I understand the signature stuff well enough for
much meaningful commentary, but the general concept sounds good.

I have one question about the backward compatibility section (3.5).
If a multi-CGA capable node needs to be able to issue ND messages
with standard SEND RSA signatures, what procedure is used to decide
when to do this?  Does the node try a Ring Signature Option, and if
it gets no response, try again with a standard SEND signature?  If
so, would it keep an notation in its neighbour cache of the Ring
Signature capability of each node, for future reference?

Cheers,
Brett.

James Kempf wrote:
> I've just posted a new draft on how to do secure address proxying. Until the
> draft appears in the Interent Drafts directory, you can find the draft here:
> 
> http://www.geocities.com/kempf42/draft-kempf-mobopts-ringsig-ndproxy-00.txt
> 
> The draft defines how to use Rivest-Shamir-Tauman ring signatures (Asiacrypt
> 2001) and CGAs generated with multiple keys to allow a router, such as a
> Mobile IPv6 home agent, to securely proxy a CGA. Multi-key CGAs allow secure
> proxying without revealing from the cryptographic material whether the
> mobile node is on or off the link, unlike other secure proxying techniques.
> 
> Note that DoCoMo has filed for IPR on this draft, we're working on the
> release currently. DoCoMo's policy is to release IPR royality-free if the
> draft is approved as an Internet Standard if the IPR involves
> interoperability, and I'm recommending that for this case.

_______________________________________________
Mobopts mailing list
Mobopts@irtf.org
https://www1.ietf.org/mailman/listinfo/mobopts