Minutes of the July 1 802.11 HT Usage Model telecon
Here follows the minutes of the July 1 802.11 HT Usage Model telecon.
A formatted version of these minutes may be found in the cumulative
minutes document, 11-03-354r4, available on the 802.11 server in
due course.
July 1, 2003
(Adrian: Many thanks to Steve Halford for taking these minutes)
Teleconference #2
Chair: Adrian Stephens, Intel Corp., adrian.p.stephens@intel.com
Secretary: Steve Halford, Intersil Corp shalford@intersil.com
Next Teleconference will be Tuesday, July 15, 2003.
Meeting called to order at 11:35 am EDT.
Agenda (from the e-mail announcing the teleconference)
1. Appoint secretary
2. Roll call
3. Modify and adopt agenda
4. Decide how many different scenarios we want
5. How should coexistence scenarios be described? Is this
in scope for this committee?
6. Create a list of the scenarios
7. Other business
Agenda Item #1: Steve Halford was appointed secretary for this meeting.
Agenda Item #2: Roll call. Participants were requested to e-mail Adrian
to confirm their attendance.
{List of attendees}
Alexei Gorokhov
Bjorn Bjerke
Christopher Hansen
Garth Hillman
Irina Medvedev
Jason Ellis
Javier Delprado
Mary Cramer
Mike Moreton
Paul Feinberg
Rahul Malik
Rolf Devegt
Sanjiv Nanda
Sean Coffey
Steve Halford
Tomer Bentzion
Vinko Erceg
Xiaolin Lu
Agenda Item #3: Agenda adopted
Agenda Item #4: Began the discussion of the number of scenarios
Rolf Devegt suggested that 5 scenarios seemed like a reasonable number
to target based on the past teleconference.
Adrian Stephens questioned if we needed separate scenarios for the 2.4
GHz band and the 5 GHz band
This was discussed within the group.
Tomer Bentzion suggested that the scenarios will be agnostic to the
band, but the channel model will address
Rolf Devegt pointed out that each band would have different regulatory
requirements that might influence the use
The group continued the discussion. It was pointed out that even within
a single band there are a variety of regulatory requirements that impact
use. General feeling of the group was that the regulatory concerns were
outside the scope of the usage cases. Some of the difference would be
important for the Channel Model subcommittee. Question was raised that
we need to be sure the usage cases and the channel model scenarios match
up sufficiently well. That is, we don't want a usage case that is
appropriate for an environment that has no channel model.
Adrian suggested we set aside the present discussion and let Vinko Erceg
review the 6 channel models being considered in the Channel Model
subcommittee.
Vinko discussed the current models from the channel model subcommittee.
(Note from the secretary: The channel model minutes can be found in
03/460r0.)
Six Models labelled A through F
Model A: Flat fading, non-line of sight (N-LOS)
Model B: Residential, line of sight (LOS) with small delay spread
Model C: Residential & small office with a mixture of LOS & N-LOS.
Larger delay spread
Model D: Typical office environment (e.g., cubicles), NLOS
Model E: Large office environment
Model F: Large open spaces, NLOS, large delay spread
Following Vinko description, the group discussed whether we needed two
usage cases for the residential environments since there are two channel
models. General consensus that the usage cases for residential
applications will use both models.
It was pointed out that we might want a channel model to cover usage of
802.11n in the large stadium, sport arena environment. Vinko felt that
Model F would cover this scenario, but we may want to revisit this and
other possible scenarios (e.g., shopping mall).
Adrian stated the position that these usage models should imply some
type of usage that is impossible with today's 802.11a/b/g systems. As
an extreme example, Adrian felt we shouldn't include something like a
single Voice over IP phone since this is something possible with today's
802.11 standards. After some discussion, there was general agreement on
this position.
Adrian reviewed where we stood with regard to the environments for the
usage models:
Residential Scenarios - Two channel models, both should be used (Model B
&C)
Two Enterprise scenarios - One medium size and one large size (Model D &
E)
Hot Spots - Could be Model E or Model F
Rolf Devegt brought up the possibility of a factory usage
environment...For example, to review and edit engineerings drawings
while on the floor of a large manufacturing building.
The sports stadium environment was also discussed
Adrian felt that we need to justify each usage model. That is, we need
to be sure that each environment and usage model was appropriate for
802.11n to address.
There was a question about channel model A and what environments/usage
model would be associated with channel model A. Vinko suggested that
Model A is meant as a general PHY model that stresses the performance.
It is not necessary for us to target a usage scenario specifically for
this channel model.
The group then began to discuss the inclusion of a coexistence model as
part of the usage model.
Matthew Shoemake reported that he has contacted 802.19 to try and get
their inputs into the 802.11n process early in the standard effort.
Adrian discussed his view of coexistence - Three cases of concern
A) Coexistence of HT and legacy 802 BSS on the same channel (do
legacy devices still function?)
B) HT AP with a mixture of legacy devices within same BSS
(fairness to legacy devices)
C) Legacy AP with a HT device associated
Adrian then expressed the opinion that good fairness for case A would be
that the throughput shouldn't decrease by more than 5%
Suggested by others that the impact in Case A should be no worse than
deploying a new 'legacy' BSS
It was also pointed out that 802.19 is chartered to look at coexistence
among all 802 devices (802.11 & 802.15 & 802.16). So, these cases that
Adrian pointed out will need to be supplemented with 802.15 devices as
well.
Group began to discuss the next steps and how to proceed in generating
the usage models.
It was suggested that we start with a list of applications. Mary Cramer
(Agere) agreed to help work on and co-ordinate gathering this list.
Others that volunteered:
Bjorn Bjerke (Qualcomm)
Paul Feinberg (Sony)
Sanjeev Sharma (Samsung)
Rahul Malik (Panasonic)
Javier Delprado (Philips)
Tomer Bentzion (Metalink)
Best Regards,
Adrian P Stephens
Intel Corporation
Tel: +44 771 276 3448 (Mobile)
Tel: +44 1223 763457 (Office)
This message was sent from the 802.19 email reflector.