| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
From: Joseph Levy
[mailto:Joseph.Levy@xxxxxxxxxxxxxxxx] Hi
Laurent and All, I
am concerned that we are using the terms STA and AP to describe the parts of an
MLD. In my view a STA, an AP, and an MLD all have a single MAC with a
single MAC SAP, hence it is very confusing to me to say that an entity
with a single MAC SAP consists of multiple entities that each have their
own MAC SAP. [RR] My concern(s) as well. We have had these discussions
in ARC for over a decade now, and the “agreement” always was (as I
recall) that a STA is a single logical addressable entity that (wirelessly) communicates
with other similar entities using a common MAC and PHY. For each PHY, there is
a unique MAC, and for each MAC, there is a unique PHY (i.e. it’s one-to-one!).
Whether or not one or more of the STAs also has a DS and is therefore
logically an “AP STA” is not relevant to the MLD discussion IMO. It seems pretty clear to me that there are only two
possibilities: 1) a single STA with a complex PHY composed of more than one
distinct band (e.g. 80+80) separated by a non-zero chunk of spectrum, however
still controlled by a single MAC that controls both “sub-bands” as
if they were a single PHY (with a single MAC address), and 2) two (multiple) STAs with 2 (multiple) MAC/PHY entities
(both (all) addressable with distinct MAC addresses) that are used “simultaneously”
by management and higher layer entities to increase the information carrying
capacity of the multiple-STA device (and yes, this does require that STA
management be coordinated by a “multiple-STA device” manager!). For
those that care, this is the approach that was taken in ISO 15 years ago (see
ISO 21217,24102-x, 21218, 1741. 17423, 17429, …) As 1) is really nothing more than a single STA with a
complex PHY (which I believe is what Joe was alluding to in his comment above …
“an AP, and an MLD
all have a single MAC with a single MAC SAP”), it is abtstractly no different
than what we have today. Thus, we are left with 2) as the “new thing”
that MLO brings to the table. I have yet to be convinced that the idea of an
upper- and a lower-MAC is the correct philosophical approach. I believe such
an approach will likely lead to “kludgy” addressing requirements, which
tells me that there is something not quite right :^))) My two cents … RR I
would change this statement to be: ·
An
EHT MLD is a physical device that contains a EHT PHY or EHT PHYs that can
operate on at least 2 links (at least 2 bands or 2 channels in one band). Regards,
Joseph From: Cariou, Laurent
<laurent.cariou@xxxxxxxxx> Thanks
all, See
some responses: To
VK: I
think the following would work •
An
EHT STA within a physical device capable of operating as an EHT STA in at least
2 bands (or 2 channels within one band) shall be part of an MLD To
Sai: Link
refers to the channel and the 2 peers (STA and AP) so not sure it’s fully
accurate, but we can use link if the group prefers. In that case that would
become •
An
EHT STA within a device capable of operating as an EHT STA in at least 2 links
(which are in 2 bands or 2 channels within one band) shall be part of an MLD To
Jay: The
intention here is just to mention that all those devices will be MLD and
therefore will support MLD functionalities that we define. Obviously the MLD
functionalities will depend on the capabilities of the devices themselves
(single radio, dual radio, STR, non-STR, …). You have to compare pre-EHT
devices and EHT devices with same capabilities (single radio, dual radio,
…), in that case the difference for EHT devices is that they’ll
support the MLD functionalities that pre-EHT device don’t. Hope
it’s clearer. To
Chunyu: We
can use link if the group prefers. What
we add here is a mandate for all devices that meet the conditions. Obviously,
if the STA wants to be dual radio, and benefit from MLD functionalities, it
will need to be supporting MLD. To
Pooya: I’ll
add that note to SP#2. To
Patrice: Obviously,
every STA affiliated with an MLD is a STA that operate with the corresponding
AP with baseline rules. So EDCA is surely supported by all STAs and we
don’t really need to mention that. Now for specific MLD devices,
especially dual radio devices that are non-STR, in addition to baseline channel
access of each STAs, we are discussing different channel access mechanisms to
improve operation (for aggregation). But this is not part of the MLO basic
framework, as this is dependent on the MLD capabilities. Thanks Laurent From: NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx> Hi Laurent, Thanks
for initiating discussion about mandatory features related to MLO features. I
have a question related to slide 3 and SP#2. I agree with you about the list of
sub features. But I think that the medium access procedure has to be defined
for the MLO. At least, a discussion must conclude on how a multi-link
transmission is initiated. If the conclusion is that standardized EDCA mechanism
is used on each link, I am afraid that a multi-link transmission can never be
started or the efficiency will be very low. So
for me, the medium access subject is a key point for the MLO. Regards. Patrice From: Cariou, Laurent <laurent.cariou@xxxxxxxxx> Hi
all, Starting
a thread for Q/A on doc 992 as we ran out of time. Feel
free to ask your questions there, and refer to the SP# to which the question
applies. Just
posted below for reference the versions of SPs for SP1 discussed this morning: •
Original SP •
An EHT STA that is capable of
operating in at least 2 bands shall be part of an MLD •
Modified SP (Brian’s
suggestion): •
An EHT STA within a device capable
of operating in at least 2 bands (or 2 channels within one band) shall be part
of an MLD •
Suggestion from Yong: •
An EHT STA shall be part of an
MLD? Thanks, Laurent To
unsubscribe from the STDS-802-11-TGBE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBE&A=1
To
unsubscribe from the STDS-802-11-TGBE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBE&A=1
To
unsubscribe from the STDS-802-11-TGBE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBE&A=1
To unsubscribe from the STDS-802-11-TGBE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBE&A=1 To unsubscribe from the STDS-802-11-TGBE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBE&A=1 |