[VOIPSEC] P2Pnat Media - A H.323 alternative to ICE

[Simon Horne]

Olivier

I guess the big design differentiator between P2Pnat Media and ICE is that 
its smarts are in the network and not the client. The network (local 
gatekeeper) decides, as part of existing call routing and before the call 
setup message is sent, how any relevant NATs are to be transversed for the 
call, removing all the complexity from the client and avoiding the need for 
the long list of "candidates" and associated media establishment delays 
generally associated with ICE.

Simon



At 04:11 PM 4/06/2007, Olivier GRALL wrote:
>Hi,
>
>If I remember, H.460.17-18-19 don't permit to optimize the media packets
>path. It's only a solution to NAT traversal routing media by a third party.
>
>An equivalent to ICE in H.323 is a very interesting thing. Simon, if I have
>enough time I will have a look to your description.
>
>Regards,
>
>Olivier GRALL.
>
>-----Message d'origine-----
>De : voipsec-bounces at voipsa.org [mailto:voipsec-bounces at voipsa.org] De la
>part de Michael Billerbeck
>Envoyé : dimanche 3 juin 2007 10:57
>À : 'Simon Horne'; voipsec at voipsa.org
>Objet : Re: [VOIPSEC] P2Pnat Media - A H.323 alternative to ICE
>
>Hi Simon,
>
>Thank you for this information.
>But there are already recommendations H.460.17, H.460.18 and H.460.19 for
>NAT traversal in H.323. Is your proposal concerning a different approach
>than H.460.x and if yes:
>Why are you using a different approach?
>Are there drawbacks in the current proposals H.460.x?
>
>Michael
>
>
>-----Original Message-----
>From: voipsec-bounces at voipsa.org [mailto:voipsec-bounces at voipsa.org] On
>Behalf Of Simon Horne
>Sent: Monday, June 04, 2007 10:50 AM
>To: voipsec at voipsa.org
>Subject: [VOIPSEC] P2Pnat Media - A H.323 alternative to ICE
>
>
>Guys
>
>This mailing has been quite for a while so I thought I'd liven things up...
>:-)
>
>A few months back I submitted a proposal to the ITU to extend H.323 NAT
>traversal capabilities to allow media to flow directly between Nated H.323
>clients rather than requiring the media to always be proxied. Basically an
>intelligent H.323 alternative to ICE
>
>An overview of the proposal and relevant documents can be found here.
>www.pacphone.com/p2pnat.html
>
>The proposal has been tested and incorporated into PacPhone windows secure
>audio/video softphone www.pacphone.com and the source code has been
>released open source and will be included as part of the GnuGk
>www.gnugk.org gatekeeper project (both of which I assist in
>maintaining).  You are free to test the proposal, both the softphone and
>the gatekeeper (windows installer (no vista), source code for linux) are
>freely available for download from the PacPhone website.
>
>If anyone has any comments and feedback (throwing knives is ok too ;-) ) I
>would be happy to discuss any aspects of the proposal.
>
>Simon
>
>
>
>_______________________________________________
>Voipsec mailing list
>Voipsec at voipsa.org
>http://voipsa.org/mailman/listinfo/voipsec_voipsa.org
>
>
>_______________________________________________
>Voipsec mailing list
>Voipsec at voipsa.org
>http://voipsa.org/mailman/listinfo/voipsec_voipsa.org