Hi all,
Once again there seems to be quite a lot of confusion about the terms we are
using. First of all, we did not make it altogether clear in the last
teleconference, what is meant by a scenario. My contribution on the issue
tried to group both the "call scenarios" and the "location management
scenarios" into one set of scenarios. We did notice during the
teleconference that this was not an adequate approach and now some of the
problems occuring in this email discussion reflect that.
For example, when Radhika is talking about starting with the intra-zone
scenario, I'm not quite sure what he is meaning. I understand that the "call
scenario" he is talking about, is the one in which both terminals are
registered to the same gatekeeper. However, this "call scenario" in itself
does not really bring anything new to H.323 unless there is something that
we need to address with the "location management scenarios". Now, while the
"call scenario" does indeed fall into the category of intra-zone
communications, the "location management scenario" for finding the address
of the B-subscriber of the call might not, i.e. the VLF and HLF that need to
be contacted may not be located in the same zone or even domain.
Furthermore, we haven't even decided that the VLF and HLF should be a part
of any zone. I know that most of us actually think that the VLF is always
located with the gatekeeper, but so far we haven't made that decision and
certainly we haven't coupled the HLF functionality always with the
gatekeeper. Thus it is quite unclear what is meant by intra-zone "location
management scenarios" and I would suggest that we do not even try to define
this term. Instead intra vs inter domain "location management scenarios"
make much more sence.
(Note that I'm using the terms "call scenario" and "location management
scenario" quite freely here. If we want to use such terms, they definitely
must be clearly defined.)
Also, I would like to point out that talking about intra-zone, inter-zone,
etc. scenarios, is actually not what was originally meant by a scenario. The
idea to introduce scenarios into the H.323 Mobility work came from Tiphon,
where 5 different scenarios are identified. These scenarios do not take into
account, whether the call (or other connection) takes place inside one zone
(intra-zone), between two zones (inter-zone) or between domains
(inter-domain). The scenarios in Tiphon simply state which network types are
involved in the connection, e.g. Tiphon scenario 1 is the VoIP-to-SCN
scenario, Tiphon scenario 2 is the SCN-to-VoIP scenario, etc. These were the
kind of scenarios I thought we were trying to identify in the last
teleconference. Of cource, each of this kind of scenarios may involve calls
that are intra-zone, inter-zone or inter-domain calls, but we shouldn't call
(no pun intended) these different call models as scenarios.
We have now chosen the all-H.323 scenario as the first one to examine
(although we didn't even know what it means). In my mind this means that:
* We only examine call cases, in which no gateways are included.
* We examine gatekeeper discovery, registration and location updating
cases, in which the user is accessing the H.323 system either in his/her
home network (e.g. domain) or a visited network (e.g. domain).
* We examine call establishment and call release cases, in which the
A-subscriber is either in his/her home network (domain) or in a visited
network (domain).
* We examine call establishment and call release cases, in which the
B-subscriber is either in his/her home network (domain) or in a visited
network (domain). Note that the call can be intra-zone, inter-zone or
inter-domain irrespective of whether the B-subscriber is in home or visited
network.
* We examine mid-call scenarios (e.g. some supplementary services), in
which the user (either A- or B-subscriber) is either in his/her home network
or in a visited network.
* In the context of H.323 Annex H we do not address the inteworking
functionalities (IWFs) between the HLF/VLF/AuF and some non-H.323 network,
unless some other document (such as H.246 Annex E) proposes additions to
H.323 Annex H. (Note that, when we start work on some other scenarios in the
future, we probably will have to take the IWFs into account.)
Other issues that need to be addressed are at least the definitions for the
home/visited network/domain/zone/gatekeeper, because it seems to me that
some members are already using some of these terms in ways that already
point to certain solutions, on which we have not yet agreed.
I will try to work out a suitable contribution for the next teleconference
to clarify these issues properly. If you have some comments to this mail,
please respond. I know that this is a long posting with quite a lot of
issues in it and I'm sure that I haven't explained myself too clearly in all
parts of it, so please ask.
------------------------------------------------
Jaakko Sundquist *
+358 50 3598281 * Audere est Facere!
jaakko.sundquist(a)nokia.com *
------------------------------------------------
-----Original Message-----
From: EXT Roy, Radhika R, ALARC [mailto:rrroy@ATT.COM]
Sent: 04. April 2000 15:31
To: ITU-SG16(a)mailbag.cps.intel.com
Subject: Re: Mob ad hoc group progress
Paul:
Thanks for clarifications.
In the last conf call, we even cannot complete the simple intra-zone
(intra-network, inter-network) scenarios first. You are right that we should
concentrate on the scenarios (Registration procedures, Gatekeeper discovery,
Location update, Call establishment, Mid-Call scenarios (e.g. supplementary
services, user interaction, Call release).
Definitely, we can go further: Inter-Zone (intra-domain) scenarios as well
if we can agree on the first one. AT&T contributions already contain many of
those aspects. Hope to bring new contributions explaining further in both
scenarios.
Best regards,
Radhika R. Roy
AT&T
> -----Original Message-----
> From: Guram Paul-LPG019 [SMTP:lpg019@EMAIL.MOT.COM]
> Sent: Sunday, April 02, 2000 3:52 PM
> To: ITU-SG16(a)MAILBAG.INTEL.COM
> Subject: Re: Mob ad hoc group progress
>
> Radhika,
>
> Conf call is for Annex H only as before...nothing has changed...change can
> only take place when the group agrees. Please do not read into the email
> more than what it says. I was only trying to highlight, in a high level
> report, the plight of Annex I since nothing much has moved in it, and it
> is
> an Annex which together with Annex H covers the mobility aspects. As far
> as
> Annex E is concerned, the point was that the mapping was the work of one
> individual or one company, thus could have errors or omissions (more than
> likely)...again only highlighting to people to check out this mapping.
>
> Contributions for next conf call to be for All H.323 intra-network
> scenario
> for:
> Registration procedures, Gatekeeper discovery, Location
> update, Call establishment, Mid-Call scenarios (e.g.
> supplementary services,
> user interaction), Call release. Also contributions on
> any
> impacts on the already specified architecture in the light of the present
> work are welcome.
>
> Paul
>
>
>
> -----Original Message-----
> From: Roy, Radhika R, ALARC [mailto:rrroy@ATT.COM]
> Sent: 01 April 2000 00:00
> To: ITU-SG16(a)mailbag.cps.intel.com
> Subject: Re: Mob ad hoc group progress
>
> Hi, Paul and Mobility Group:
>
> I like to see that it should be clarified via emails
> whether
> H.323 Annex I
> and H.246 Annex E will be a part of the conference call.
> My
> personal
> preference is not to discuss annexes I and E during the
> conference call
> although email discussions will be preferred.
>
> If Annex I and E are included in the conference call, I
> like
> to see the
> actual time in the agenda when these items will be
> discussed
> so that people
> can best use their time in joining the particular time
> slot
> for each Annex.
>
> The proposal is as follows:
>
> 1. The upcoming conference call to be dedicated for H.323
> Annex H.
> 2. Let both editors of H.323 Annex I and H.246 Annex E
> propose via emails
> whether any issues to addressed.
> 3. If annex I and E are included in the agenda of the
> upcoming conference
> call, I like to the time slots when these annexes will be
> discussed so that
> people can join the particular time slots of interest.
> Alternatively,
> separate conference calls can be arranged for annexes I
> and
> E.
> 4. Are editors or any members of the mobility group sure
> what would be the
> scope of work (I guess that Annex E is stable for now -
> thanks to the
> editor) for those annexes?
>
> Paul: Would you please clarify what specific area that we
> need to discuss
> for the next conference call for H.323 Annex H? My guess
> is
> that we may
> start with intra-zone communication first. I am planning
> to
> bring
> contribution(s) describing this for mobility management:
> Discovery,
> Registration, Location Updates, and Call Establishment.
>
> I would appreciate comments form all members.
>
> Best regards,
>
> Radhika R. Roy
> AT&T
> H.323 Ad Hoc Mobility Group
>
