Hi,

Not sure I understand exactly what you want to do, but if you want your 
station to authenticate an AP, you could actually use 802.1X with the roles 
reversed: the client would be the authenticator (also action as the auth 
server), and the AP would be the supplicant. Depending on the exact setup, 
this might be more or less easy to do (it will require having the code for 
the EAP method used on the boxes themselves).

Note that cisco APs can act as LEAP clients in some situations (when you 
have several APs, one of which is a repeater -acting as a supplicant- and 
the other being the root -acting as a regular authenticator- for instance). 
Standard 802.1X is not supported in this case, though (LEAP is limited to 
one single and pretty simple method).

This is the normal setup:

Station            AP                          RADIUS server
(supplicant)       (authenticator)             (auth server)
EAP method(s)      no EAP methods              EAP method(s)
(client side)      (transparent passthrough)   (server side)
client credentials                             list of clients with 
verifiers for each

This is, I believe, the setup you're looking for:

Station               AP
(authenticator)       (supplicant)
EAP method(s)         EAP method(s)
(server side)         (client side)
list of APs with      AP credentials
verifiers for each

Of course, that means that each station has to carry a copy of the 
verifiers (depending on the method these can be passwords, encrypted 
passwords, root certs, CRLs, etc.) for all APs, and get an update when the 
list changes. Alternatively, the station could actually contact a RADIUS 
server to verify the info, but that means using the link that has not been 
verified yet (unless of course you have another link somewhere else). Using 
IPsec to protect the RADIUS conversation might not be a bad idea in this 
case...

Jacques.

At 00:02 09/10/2002, David Rhodes wrote:
>Thanks to all for the replies on this one, especially the outline below.
>  A bit of additional info - I do use VPN's for connection security, so I'm
>not really worried about MITM attacks via rogue AP's. I have a different
>application I'm working on related to mesh networking where I need to
>identify the AP for QOS reasons. A standard smart card empowered
>protocol/API would indeed do the trick, yet it seems fairly obscure in
>actual implementation at this point.
>Is there maybe a hack that could be done to extract the EAP (or equiv.)
>identifier of the authenticator over the wireless link? (by a client posing
>as an auth server) Or does 1X/EAP-TTLS forbid auth traffic over the air -
>i.e. backhaul only? Very curious.
>
>Thanks again..
>
>David
>
>-----Original Message-----
>From: Fors Chad-QA3336 [mailto:[EMAIL PROTECTED]]
>Sent: Tuesday, October 08, 2002 12:29
>To: 'David Rhodes'; [EMAIL PROTECTED]
>Subject: RE: [BAWUG] AP spoof detection
>
>
>802.1X provides the framework for the authentication, but the actual
>procedure (and whether or not the it supports mutual authentication) depends
>on the EAP method chosen.  The simplest method is EAP-MD5 challenge, which
>authenticates the user to the authentication server, but not the
>authentication server to the user.  More advanced methods (e.g., EAP-TLS,
>EAP-TTLS, EAP-SIM, etc.) provides for mutual authentication, and is fully
>compatible with 802.1X.  These methods will authenticate the authentication
>server to the client, but will NOT authenticate the AP (i.e., authenticator)
>to the client.  Therefore, if one of these methods is used, the following
>relationships should result:
>
>Nodes are:
>         S  = Supplicant (i.e., client, e.g., laptop)
>         A  = Authenticator (e.g., AP)
>         AS = Authentication Server (e.g., RADIUS Server)
>
>o S has authenticated AS via EAP method
>o AS has authenticated S via EAP method
>o A has authenticated S via 802.11 authentication (e.g., Open Auth)
>o AS has "authenticated" A via RADIUS _RequestAuthenticator_ in RADIUS
>request message
>o A has "authenticated" AS via RADIUS _ResponseAuthenticator_ in RADIUS
>response message
>o S has NOT authenticated A explicitly
>
>Therefore, even though S has not authenticated A explicitly, S has
>authenticated AS, who in turn has authenticated A, so their may be an
>implied trust between S and A.  To summarize, then, if the client can
>authenticate the server (via an EAP method that supports mutual
>authentication), it essentially can trust the Access Point in between.
>
>Let me know if I confused the issue even more, and I will attempt to clarify
>:o)
>
>-Chad Fors
>
>     He does not believe, that does not live according to his
>belief.
>                                   ... Thomas Fuller (1608-1661)
>
>
>
>-----Original Message-----
>From: David Rhodes [mailto:[EMAIL PROTECTED]]
>Sent: Tuesday, October 08, 2002 11:29 AM
>To: [EMAIL PROTECTED]
>Subject: [BAWUG] AP spoof detection
>
>
>..another thought related to recent EAP/LEAP threads - Does anyone know if
>any of the related 1x mechanisms will provide AP authentication to the
>client? It seems like all the effort has gone into authenticating the
>client, not the access point. I realize that most 802.11 equip. was built
>for corporate and home environments where the network provider is trusted,
>but this is not true in the public space.
>  I haven't used the 1x solutions to any serious degree yet but it appears
>the AP only passes the supplicant info to the RADIUS server. I know the
>RADIUS server essentially auth's the AP via the optional SSL/shared key
>connection but that doesn't provide the user any first hand information.
>Seems like we need some way to put public certs on the AP's similar to what
>is done with webservers. With all these stories of pimping starbucks wifi
>customers from the street, etc..not to mention AP storms...  or am I missing
>something?
>
>thanks,
>david
>
>--
>general wireless list, a bawug thing <http://www.bawug.org/>
>[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
>
>--
>general wireless list, a bawug thing <http://www.bawug.org/>
>[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless


-- Jacques Caron, IP Sector Technologies
    Join the discussion on public WLAN open global roaming:
    http://lists.ipsector.com/listinfo/openroaming

--
general wireless list, a bawug thing <http://www.bawug.org/>
[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless

Reply via email to