Re: [Mobopts] Key transfer issue for pana-cxtp

From: James Kempf <kempf(at)docomolabs-usa.com>
Date: Fri Apr 01 2005 - 13:58:33 EST

> > > From this AAA-Key-int, the nPAA computes a new AAA-Key using nonces
> > > exchanged in PANA-Start-Exchange between MN and nPAA
> > >
> > > AAA-Key-new = The first N bits of
> > > HMAC-SHA1(AAA-Key-int, PaC_nonce | PAA_nonce)
> > >
> >
> > Is the exchange between PaC and nPAA in which PaC_nonce is send
protected? I
>
> the PANA-Start-Exchange (PSR/PSA) is not protected
>

Then I doubt this would pass muster. An attacker that had compromised the previous PAA could snoop the start exchange and reconstruct the key. Unlikely, but nevertheless.

> > I'd suggest you get a copy of the Housley Critera and run through a
proof
> > that they are satisfied, if you really want to make sure.
>
> I got it, The criteria of interest is:

Like I said above, I am doubtful this key exchange protoocol will pass. I don't think this statement is meant to exclude the possibility that other actions in addition to compromising one NAS might be required.

            jak

---

Mobopts mailing list
Mobopts@irtf.org
https://www1.ietf.org/mailman/listinfo/mobopts Received on Fri Apr 1 13:59:06 2005