All (anyone with an opinion),
So, can I confirm then that we have a consensus that we can dispense with the definition of ESA, and we can use the below to remove the ESA concept from the text? That was the intention of the comment, and Stephen’s resolution, if the below body text update resolves it, that would be great.
Mark
From: Perahia, Eldad <eldad.perahia@xxxxxxx>
Sent: Tuesday, March 15, 2022 4:40 PM
To: Mark Rison <m.rison@xxxxxxxxxxx>; mark.hamilton2152@xxxxxxxxx; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
Hi Mark R.,
Your suggestion is acceptable to me. Thanks.
Regards,
Eldad
From: Mark Rison <m.rison@xxxxxxxxxxx>
Sent: Tuesday, March 15, 2022 2:47 PM
To: Perahia, Eldad <eldad.perahia@xxxxxxx>; mark.hamilton2152@xxxxxxxxx; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
I think the issues are the "blanket coverage" and the "several BSSs".
I suggest:
The Planned ESS subfield indicates whether the BSS is part of an ESS that has at least two BSSes whose coverage overlap.
If it isn't, this subfield is set to 0 and the Edge Of ESS and Recommended BSS Transition RSSI Threshold Within ESS subfields are reserved.
If it is, this subfield is set to 1, and:
- The Edge Of ESS subfield indicates whether the BSS is at the edge of the ESS. This subfield is set to 1 to indicate the BSS is at the edge of the ESS. Otherwise, this subfield is set to 0.
- The Recommended BSS Transition RSSI Threshold Within ESS subfield indicates the RSSI below which an associated STA is recommended to initiate BSS transition to a neighbor BSS belonging to the ESS.
Thanks,
Mark
--
Mark RISON, Standards Architect, WLAN English/Esperanto/Français
Samsung Cambridge Solution Centre Tel: +44 1223 434600
Innovation Park, Cambridge CB4 0DS Fax: +44 1223 434601
ROYAUME UNI WWW: http://www.samsung.com/uk
From: Perahia, Eldad <eldad.perahia@xxxxxxx>
Sent: Tuesday, 15 March 2022 21:12
To: mark.hamilton2152@xxxxxxxxx; Mark Rison <m.rison@xxxxxxxxxxx>; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
I don’t see the confusion with terms “Planned ESS” and “Edge of ESS”, so its hard for me to come up with alternative suggestions.
If the main issue is the phrase “blanket coverage” and use of ESA, we can delete them as follows:
The Planned ESS subfield indicates whether the BSS is part of an ESS that is planned with several BSSs in an overlapping configuration, and if so the subfield is set to 1. This subfield is set to 1 to indicate that the ESS is deployed to ensure blanket coverage over the Extended Service Area (ESA). Otherwise, this subfield is set to 0 and the Edge Of ESS and Recommended BSS Transition RSSI Threshold Within ESS subfields are reserved.
If this doesn’t work, may I suggest that someone provide a rewrite, and I’ll review to see if I agree that it meets the original intent?
Regards,
Eldad
From: mark.hamilton2152@xxxxxxxxx <mark.hamilton2152@xxxxxxxxx>
Sent: Tuesday, March 15, 2022 10:08 AM
To: Perahia, Eldad <eldad.perahia@xxxxxxx>; 'Mark Rison' <m.rison@xxxxxxxxxxx>; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
Eldad,
Two thoughts:
- First to clarify. I don’t support/suggest deleting features, here. Just simplifying (?) the way we describe the features, to say in-line what the “Planned ESS” and “Edge of ESS” mean, without the ‘hassle’ of an ESA definition, and further a definition that isn’t actually referenced when the “Edge of ESS” concept is specified, by the way.
- We seem to be blurring the concepts of “Planned ESS” and “Edge of ESS”. I’m actually not sure how these concepts are each supposed to relate to the ESA, and the question of whether the ESA can have “holes” (and are such holes an “Edge”). It is very interesting to note that the definition of ESA clearly/explicitly mentions that it might be disjointed. So, it seems a Planned ESS (which does mention ESA explicitly) can have holes, despite the use of “blanket coverage” in the description. In fact, my reading of the definition is that an ESA might not even be completely connected – it might not only have “holes” but it might have “islands”. This leads me to thinking the phrase “blanked coverage” is misleading – that blanket is a blanket with both holes and rips into pieces, per definition – which might not be intuitive to the reader who doesn’t double-check clause 3.
So, I think it would be very good (necessary) to clarify in 9.4.2.256 and 11.21.7.5 that the “Planned ESS” concept includes having holes/islands in the coverage. And, further to clarify how the “Edge of ESS” relates to these holes and areas of disjointedness (or any concept of ESA, if we keep that concept/term). I really don’t care whether we do that by referencing the definition in more places in the text (and probably re-enforcing these concepts in the text as well), or just by describing all this directly in the text in words instead of a definition reference.
Mark
From: Perahia, Eldad <eldad.perahia@xxxxxxx>
Sent: Tuesday, March 15, 2022 10:19 AM
To: Mark Rison <m.rison@xxxxxxxxxxx>; mark.hamilton2152@xxxxxxxxx; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
I think we need to be extremely careful with deleting features that are in current use in industry.
The main point of this feature is for the planned/managed network to assist non-AP STAs in roaming. And to furthermore indicate when they are leaving said area. Think stadium or shopping mall or office building where transitions between BSS’s are designed as best as possible to a specific RSSI and absence of coverage when leaving the ESS/ESA.
There is no implied guarantee of “no coverage holes”, as such a guarantee is pretty much impossible with the randomness of RF propagation. I’m not clear on the background of the comment, but if the issue is with “ensure blanket coverage”, you could change the wording to “…the ESS is deployed to provide coverage to the best of our ability over the ESA…”.
Or is the issue with ESA? This I don’t understand since in 802.11-2020 we have “extended service area (ESA): The area within which members of an extended service set (ESS) can communicate. An ESA is larger than or equal to a basic service area (BSA) and might involve several basic service sets (BSSs) in overlapping, disjointed, or both configurations.” Even in the definition of ESA, they use the term “disjointed”, so no guarantee of perfect coverage.
As such, I think the original wording covers the intent of feature clearly enough.
Regards,
Eldad
From: Mark Rison <m.rison@xxxxxxxxxxx>
Sent: Tuesday, March 15, 2022 1:59 AM
To: mark.hamilton2152@xxxxxxxxx; STDS-802-11-TGM@xxxxxxxxxxxxxxxxx; Perahia, Eldad <eldad.perahia@xxxxxxx>
Subject: RE: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
> I believe Mark R is trying to get to the idea there are no physical area gaps between the BSAs. 1) I’m not sure how to say that (other than some term like convex hull – although I’ve never heard convex hull applied to shapes, only to points); and 2) I’m not convinced that we have a consensus understanding that the ESA was meant to imply such “non-disconnected” coverage area.
> I’m inclined to stick with just deleting this concept. I know Mark R disagrees.
[I'm not a mathmo, so I may well be using "convex hull" incorrectly.
An alternative could be "The combination of the BSAs is topologically
equivalent to a sphere, i.e. no holes."]
The problem is that if you delete "Planned ESS" then the remaining fields
in the (ESS Report) element are reserved, and hence there's no point
having the element at all.
I think the intent was that if you're in a "Planned ESS" then you
can find out when you're about to walk out of the ESS (because the
Edge Of ESS subfield is set). This doesn't work so well, in general,
if there are "holes" in the ESS coverage.
I suppose you could argue that it's OK if there a "holes" as long
as all the BSSes that are not completely surrounded by other BSSes of
the same ESS advertise they're at the edge. In that case maybe the
rules would be something like:
The Planned ESS subfield indicates whether the BSS is part of an ESS that is planned with several BSSs in
an overlapping configuration. This subfield is set to 1 to indicate that the ESS is deployed to ensure
that a STA can move from any BSS in the ESS to any other BSS in the ESS while remaining in the
BSA of at least one BSS in the ESS at all times. Otherwise, this subfield is set to 0 and the Edge Of ESS
and Recommended BSS Transition RSSI Threshold Within ESS subfields are reserved.
The Edge Of ESS subfield indicates whether the BSS is at the edge of the ESS. This subfield is set to 0 to
indicate the BSS is not at the edge of the ESS, i.e. that the BSA overlaps with the BSA of another BSS of
the ESS in every direction. Otherwise, this subfield is set to 1.
? That gets rid of the term "ESA" too.
I vaguely recall that Eldad presented this (although searching Mentor
doesn't support this recollection), so maybe he can help.
Thanks,
Mark
--
Mark RISON, Standards Architect, WLAN English/Esperanto/Français
Samsung Cambridge Solution Centre Tel: +44 1223 434600
Innovation Park, Cambridge CB4 0DS Fax: +44 1223 434601
ROYAUME UNI WWW: http://www.samsung.com/uk
From: Mark Hamilton <mark.hamilton2152@xxxxxxxxx>
Sent: Monday, 14 March 2022 21:10
To: STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
--- This message came from the IEEE 802.11 Task Group M Technical Reflector ---
Stephen,
I think it is a tautology that the ESS covers the area that is the union of the BSAs. I would not want to say that (again), and confuse anybody that it wasn’t already the case.
I believe Mark R is trying to get to the idea there are no physical area gaps between the BSAs. 1) I’m not sure how to say that (other than some term like convex hull – although I’ve never heard convex hull applied to shapes, only to points); and 2) I’m not convinced that we have a consensus understanding that the ESA was meant to imply such “non-disconnected” coverage area.
I’m inclined to stick with just deleting this concept. I know Mark R disagrees.
Mark
From: Stephen McCann <mccann.stephen@xxxxxxxxx>
Sent: Monday, March 14, 2022 3:05 PM
To: STDS-802-11-TGM@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-11-TGM] CID 2315 re-consideration - was [Pull requests for today's motions]
--- This message came from the IEEE 802.11 Task Group M Technical Reflector ---
Mark,
thanks for your work on this CID 2315:
CID 2315 in MAC-AL: I don't think this is correct. I think the point of "ESA" (albeit perhaps not clear) is that if you have Planned ESS set then you know there are no "dead spots" within the area covered by the ESS (i.e. topologically there are no holes). Maybe the wording should be closer to "This subfield is set to 1 to indicate that the ESS is deployed to ensure that there is coverage over the entire area bounded by the convex hull of the BSAs of the BSSes in the ESS"
I think your revised wording is ok, but I would prefer not to introduce "convex hull" as a new topological term into the specification. I think alternative wording could be "This subfield is set to 1 to indicate that the ESS is deployed to ensure that there is coverage over the entire area bounded by the BSAs of the BSSes in the ESS".
Do you have any further comments? Thanks
Kind regards
Stephen
To unsubscribe from the STDS-802-11-TGM list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGM&A=1
To unsubscribe from the STDS-802-11-TGM list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGM&A=1
To unsubscribe from the STDS-802-11-TGM list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGM&A=1