RE: [Mobopts] New topics for the RG
Colleagues,
The following draft is of interest to the IEEE 802.21 community and I
encourage everyone to be familiar with the concepts.
http://www.ietf.org/internet-drafts/draft-daniel-mip-link-characteristic
-02.txt
Please notice in particular the implications of defining IEEE 802.21
parameterization of link change events and possible re-parameterization
of them in another layer. Coordination between groups and goals would be
useful here.
Also, the suggestion of combining link event reporting with a MIP-based
protocol has been discussed in our forum.
Best Regards,
Michael G. Williams, IEEE 802.21 Vice Chair
________________________________
From: mobopts-bounces@irtf.org [mailto:mobopts-bounces@irtf.org] On
Behalf Of ext Rajeev Koodli
Sent: Wednesday, July 13, 2005 10:52 AM
To: Ji Zhang
Cc: Soohong Daniel Park; mobopts@irtf.org
Subject: Re: [Mobopts] New topics for the RG
Hi,
I have seen interest in this topic. However, I think the topic is
more than just end-to-end event notification (Disclaimer: I have not
read the latest version). It could affect traffic on the Internet.
So, before we could go ahead with the solution(s), we need to
carefully look into the broader problem, including "cross-layer
design" implications, congestion-control etc. Would you be
interested in looking into those first?
Regards,
-Rajeev
ps: A while ago, MIP group looked into the problem of end-to-end
QoS problem definition and requirements. I feel the problems are
similar.
Ji Zhang wrote:
Hi Rajeev,
Thanks a lot for your summary.
We would like to recommend the following topic to be considered
for the
charter:
Deliver link layer characteristic information (e.g.
bandwidth...) to a
mobile node's communication peers using mobility control
messages (i.e.
binding updates or registration requests), so that the
communication peers
of the mobile node can control their traffic flows accordingly.
The idea is
based on the draft
http://www.ietf.org/internet-drafts/draft-daniel-mip-link-characteristic
-02.txt.
With this approach, the potential problems (i.e. capacity
overspending or
underutilization) on access link capacity changes (especially on
vertical
handovers) can be solved. The proposal is suitable for both
MIPv4 and MIPv6,
regardless of the type of the transport protocol used.
Simulations will be
done in the near future to test its performance.
Any comments and questions are welcome.
Regards,
Ji
Now, it is time to re-consider our charter items. What
sorts of topics
are folks interested in working on? I would like to
broaden our scope to
consider Mobility on the Internet in general. Some
possible topics for
discussion below in no preference order. Please comment,
suggest your
topics and provide a brief justification.
- multicast and mobility. A specific problem is what
address should the MN
use as its source IP address so that its packets are
not discarded due
to a) RPF checks,
and b) source-specific multicast. To pass a), you need
the MN to use
CoA, but to
pass b), you need HoA. How to convice multiple
receivers about CoA and
HoA binding?
Is RR even an option? :-)
- mobility (and traffic) patterns in a WLAN (e.g., a
campus network).
This should provide
a better sense of traffic for admission control as
well as
network-controlled handovers.
- what is the scope of IP paging in a converged WLAN -
WWAN environment?
Is it a real problem? Worthwhile documenting in any
case?
- architectural barriers to optimizing inter-domain
handovers. Although
we have
some understanding of improving delay and packet loss
during
handovers across
IP networks, how applicable are they across different
autonomous
systems? Would
the policy barriers allow optimizations? Is this an
example of
"tussle in cyberspace"?
- privacy topics related to mobility
- Others
-Rajeev
------------------------------------------------------------------------
----
----
_______________________________________________
Mobopts mailing list
Mobopts@irtf.org
https://www1.ietf.org/mailman/listinfo/mobopts
_______________________________________________
Mobopts mailing list
Mobopts@irtf.org
https://www1.ietf.org/mailman/listinfo/mobopts