Re: Example of MeGaco messages for Multiplex terminations

"Schwarz, Albrecht (Albrecht)" <[email protected]> Fri, 14 Oct 2011 14:45:23 +0200
Newsgroups gmane.ietf.megaco
Message-ID <5F7BCCF5541B7444830A2288ABBEBC962160D26A45@FRMRSSXCHMBSD2.dc-m.alcatel-lucent.com>
Question for clarifications:
the right hand side seems to be not any multimedia-over-H.223 service (due to separate ephemeral terminations for audio & video), rather multimedia-over-packet like H.323 or SIP.

Is my assumption correct that the left hand side relates to a protocol stack like 
{Audio, Video}-over-H.324/H.226/H.223-over-N x Channels, right?

Thus, there would be then two levels of multiplexing, H.226 & H.223, right?

You did select for each a Mux termination, which logically belong both to the left side of the Context. 
[The right hand side should look like Figure 6/H.248.1, i.e., there wouldn't be any Mux termination at the right hand side.]

It looks like that your use case is addressed by H.248.1, § 6.2:
"Multiplexing terminations may be cascades (e.g., H.226 multiplex of digital channels feeding into a H.223 multiplex supporting an H.324 session)."

Thus, cascades are allowed, but I'm not really familiar with the MuxType element.

Christian,
do you know the syntax for creating Mux terminations with hierarchical multiplexing levels?

BR,
Albrecht


> -----Original Message-----
> From: [email protected] 
> [mailto:[email protected]] On Behalf Of Javi Muñoz
> Sent: Freitag, 14. Oktober 2011 14:00
> To: [email protected]
> Cc: Claudio Fontana
> Subject: Re: [Megaco] Example of MeGaco messages for 
> Multiplex terminations
> 
> El 14/10/2011 13:27, Claudio Fontana escribió:
> > My answer might be OT for this list, but remember that if you are 
> > using megaco for voice in a 3gpp network, multiplexed 
> terminations are 
> > not supported in any of the 3gpp releases.
> >
> > ref: 29.232 C.6.2 "Multiplexed terminations"
> >
> In my case (supporting of Nx64 multichannel calls), these 
> terminations are the only solution. So, I need to build de 
> MeGaco messages to create them.
> 
_______________________________________________
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Javi Muñoz
Sent: Freitag, 14. Oktober 2011 12:29
To: [email protected]
Subject: [Megaco] Example of MeGaco messages for Multiplex terminations

Hi,

I have read that a multiplexing termination hides all of the terminations it contains from the rest of the context, so the streamIDs of the streams flowing between the contained terminations and the multiplexing termination are local rather than global in scope.

If e have the following situation:

                         Context interior ---->
                     _________                 _________
        Data stream  | H.226 |                 | H.223 |   Audio stream
Ckt 1 ..............| "MUX" |                 |  MUX  | .................
        . . .        | term  |   Data stream   | term  |
        Data stream  |       | ................|       |   Video stream
Ckt n ..............|       |                 |       | .................
                     |       |                 |       |
                     ---------                 ---------

How we know that the stream between the H.226 and H.223 multiplexing terminations is a data stream, while the streams passing from the H.223 termination to the rest of the context are audio and video?. I think we need an explicit description of the exterior-side and context-side streams supported by each multiplex somewhere within our Megaco/H.248 documentation but, how may it be indicated in a MeGaCo meesage?

Also, if we want to use separate RTP terminations for audio and video, I need to be able to tell that mux stream 1 shall be mapped to RTP termination T3, and stream 2 into RTP termination T4. How may it be indicated?


In summary, may someone propose the full MeGaCo messages to create this scenario with multiplex terminations?

_______________________________________________
Megaco mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/megaco>