eap Mailing List: Re: EAP-AKA Key derivation

[Thomas Otto (t.otto]

Hi Bernard, all,  

> It is very common for IETF specifications to reference cryptographic
> functions defined outside of the IETF.  Examples include AES and DES, both
> of which are defined in NIST documents.

This may be, but in the IETF, also appropriate mechanisms have been developed.
As I said before, for instance the prf+ in IKEv2. 

So why abstain from using this PRF? The HMAC construction seems much more
comprehensible than the DSS-PRF. 

Is there a particular advantage which I haven't recognized? IS the DSS-PRF 
built-in in the USIM, for instance? This would be an argument, to re-use the
implementation.