Re: IAPP on hostAP


From: Jacques Caron (Jacques.Caron_at_IPsector.com)
Date: 2002-08-17 16:31:42 UTC



At 22:38 16/08/2002, Jun Sun wrote:
>On Wed, Aug 14, 2002 at 03:30:48PM +0200, Jacques Caron wrote:
> > The 802.11f draft is available in the members-only area of the IEEE 802.11
> > working group... :-/
> >
>
>Is this supposed to be a secret? Or someone who have access can re-post
>it somewhere?

I think reposting them is not allowed. Many people have criticized the lack of openness of IEEE, but even though it is slowly changing, those documents are still members-only afaik. You might want to ask the chair of the WG if you can get it, the info should be somewhere in the 802.11 pages... Don't know what the real official policy is (I'm not a member of any of those WGs).

> > Note that if you use layer 2 roaming, Mobile IP should not see/have to do
> > anything. Mobile IP should only need to do anything if you move to a
> > different network. It would be interesting to compare performance of L2
> > handoffs vs. Mobile IP handoffs (with all the back-end AAA stuff etc.),
> though.
>
>If I understand it correctly, Mobile IP requires a home base server,
>while L2 uses vlan. It seems to me L2 approach should be faster,
>because it only needs to broadcast a message so that switch can
>update its table, while in the case of mobile IP there should a
>series of actions (obtaining new IP for mobile client, mobile
>client sending packets back to home base server, etc)

If you exclude AAA, yes, the L2 approach should be faster. But moving to another AP should imply some AAA (if you use 802.1X to authenticate access), which is not necessarily shorter in number of roundtrips... Note that in the case of Mobile IP, the IP of the mobile node does not change (it's the whole purpose of Mobile IP), but indeed one has to communicate to the Home Agent the new location of the mobile node.

Jacques.



This archive was generated by hypermail 2.1.4.