Re: [MEXT] Well-known problem with authentication/etc. in wireless networks

[Pete McCann <[e-mail]>]

Hi, Julien,

On Thu, Aug 25, 2011 at 6:39 PM, Julien Laganier <[e-mail]> wrote:
>
> In the context of this discussion, optimizing Mobile IPv6 handover
> speed seems to imply that slowness of those is the root cause of that
> lack of MIPv6 deployment, which I don' t think is the case. On the
> other hand, coupling network access authentication with mobility
> management would arguably reduces deployment flexibility and thus harm
> rather than help potential MIPv6 deployments.

You're right that performance is probably not why 3GPP went the
way they did.  I think there were a variety of motivations that boil
down to a desire on the part of operators to control the handling
of UE traffic, mainly by routing it back through policy enforcement
points.  They built a network-based mobility management scheme
that borrowed very little from Mobile IP.

There was also a legacy authentication infrastructure built around
AKA that they wanted to re-use.  They didn't build algorithm agility
into their UE authentication protocol.  One could argue that in an
LTE network the mobility is very much tied to the access authentication,
and it's one reason why MIPv6 has missed the boat here.

> Thus I am still not sure what the problem is.

There's probably very little impetus for change no matter what MEXT
does.

-Pete