Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [STDS-802-11-TGBE] Clarification of CID 14115



Thanks Minyoung, Zhou for the discussions and clarifications.

 

Now my thinking is we are very obsessed with the rules defined in EMLSR but changing rules defined in sounding protocols. IMO, making the sounding protocol unique if EMLSR STAs present doesn’t seems to be a decent design.

 

 

BRs,

Xiaogang.

 

From: Minyoung Park <mpark.ieee@xxxxxxxxx>
Sent: Thursday, December 15, 2022 3:09 PM
To: Xiaogang Chen <xiaogang.chen@xxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-11-TGBE] Clarification of CID 14115

 

Hi Xiaogang,

 

The current rules in the spec is that once a STA in EMLSR detects a frame that is not for the STA, it will start to transition back to the listening operation and that transition could happen in the middle of the 1st BFRP frame so the diagram you have wouldn't happen.

 

Regards,

Minyoung 

 

On Thu, Dec 15, 2022 at 12:01 PM Xiaogang Chen <xiaogang.chen@xxxxxxxx> wrote:

Agree with Zhou.

Either we add exception on top of the new rule in sounding subclause, or in the EMLSR subclause add:

  1. AP should include all EMLSR STAs in the 1st BFRP..
  2. Non-AP STA operate in EMLSR shall fall back to listening if the STA is not addressed in the 1st BFRP…

 

For the case I was concerned about. In figure below, if the AP doesn’t get any response from the 1st BFRP, can Ap send the 2nd BFRP for EMLSR STA sifs after the 1st BFRP? EMLSR STAs need a transition time to switch back to listening, if the transition time is much larger than SIFS. The case in the figure seems feasible.

 

BRs,

Xiaogang.

 

From: Zhou Lan <z_lan@xxxxxxxxx>
Sent: Thursday, December 15, 2022 11:39 AM
To: Xiaogang Chen <xiaogang.chen@xxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-11-TGBE] Clarification of CID 14115

 

Thanks Xianggang for raising this up. With this new rule added. It looks not correct to just simply reject the comment. On one hand, a beamformee after seeing its AID in the NDPA will expect to receive a BFRP while on the other hand, a beamformee that is under eMLSR will switch back to listen mode if the first BFRP is not addressing to it and will not wait for a later BFRP. 

 

IMO, some clarification is needed on both AP and STA sides. 

 

Thanks

 

Zhou 

 

On Dec 15, 2022, at 11:27 AM, Xiaogang Chen <xiaogang.chen@xxxxxxxx> wrote:

 

In addition, we have the rule below added in 1871 which makes non-AP MLD more confused when a STA associated with the non-AP MLD operating in EMLSR mode…if the STA is not pulled in 1st BFRP should the STA wait till the end of TXOP and get prepared for another BFRP? Or go back to listening? I guess it’s the later one?

 

Another case: 4 STAs in total, AP pull 2 STAs with 1st BFRP but doesn’t get any response within SIFS (due to interference or whatever), Can the AP pull the other two STA which are in EMLSR mode? Note that 2nd BFRP is only about SIFS away from the endo of 1st BFRP.

 

(#11671) In the EHT TB sounding sequence, the STAs identified in the NDP Announcement frame should be the same as the ones identified in the Trigger frame(s) in the same TXOP.

 

BRs,

Xiaogang.

 

From: Xiaogang Chen <xiaogang.chen@xxxxxxxx> 
Sent: Thursday, December 15, 2022 10:28 AM
To: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-11-TGBE] Clarification of CID 14115

 

Hi Minyoung,

I understand the CID below passed SP this morning. But my question is still hanging... Maybe you or Gauran can help to clarify.

Your reason of rejection is a clarification. Why not added into the spec to clarify? I understand this reason may be derived from existing text implicitly. But the behavior of sounding feedback from EMLSR STA is indeed different from a STA not in EMLSR mode. For instance, after receive NDPA/NDP, a STA will prepare the feedback. If the STA is not addressed in the 1st BFRP, a non-EMLSR STA will keep the feedback in memory till a potential BFRP addressed to this STA. but if the STA is in EMLSR mode, it will flush the memory and go back to listening. 

It would be helpful to clarify at least from the non-AP STA side. Thanks.

 

14115

Li-Hsiang Sun

35.3.17

464.30

"The STA affiliated with the non-AP MLD that received the initial Control frame does not
respond to the most recently received frame from the AP affiliated with the AP MLD that
requires immediate response after a SIFS"

This should not be applicable to sounding because EMLSR STA may not be the first group of STAs triggered by the first BFRP trigger

For sounding sequence, the end of frame exchange does not occur until the EMLSR receives a BFRP trigger or until intra-BSS NAV expires

Rejected

 

The EMLSR switch back rules are quite clear - an AP knows that if it does not schedule all the STAs affiliated with non-AP MLDs in EMLSR mode (that it has triggered via BSRP/MU-RTS and included in the NDP Announcement frame) in the first BFRP Trigger frame, the non-AP MLDs that aren’t included in the first BFRP Trigger frame will switch back to listening operation.

 

 

 

 

BRs,

Xiaogang.

 

From: Minyoung Park <mpark.ieee@xxxxxxxxx> 
Sent: Friday, December 2, 2022 10:37 AM
To: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-11-TGBE] Initial Revision of Agendas for Nov'22-Jan'23 Telcos Uploaded

 

Hi Alfred,

 

Please add doc 11-22/1848r2 (1 CID for SP) to the MAC queue.

- CID 14115

 

Thanks,

Minyoung

 

 

 

On Mon, Nov 28, 2022 at 1:15 PM Alfred Asterjadhi <asterjadhi@xxxxxxxxx> wrote:

Hello all,

 

I uploaded the telco agendas for the Nov'22-Jan'23 time period. 

 

Please take a look at the submissions list and let me know if any of the contributions are missing from the queues. 

 

In addition please take a look at the contributions that are listed for the first and second MAC (this weeks) teleconferences, and let me know if there is a need for any updates.

 

Last but not least, please make sure that if you have a submission with a request for SP then that submission appears in the queue as pending SP. This way it does not end up in the quarantine queues/docs. Updated revision of docs containing CIDs in quarantine will be uploaded by end of this week.

 

Best Regards,


Alfred

 

-- 

Alfred Asterjadhi, PhD

IEEE802.11 TGbe Chair,

Qualcomm Technologies Inc.

Cell #:    +1 858 263 9445

Office #: +1 858 658 5302


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


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