Re: [STDS-802-11-TGBE] Discussion on power save of AP MLD

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

Hi Morteza,

Thanks for your discussion. Please find my response inline.

Regards

Guogang Huang

发件人: Morteza Mehrnoush [mailto:mmehrnoush@xxxxxx]
发送时间: 2022年4月1日 9:31
收件人: Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; huangguogang <huangguogang1@xxxxxxxxxx>; 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
抄送: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
主题: Re: Discussion on power save of AP MLD

Hi Guogang,

Thanks for this contribution. I have a few questions/comments:

* In the current spec, for a STA to go to power save mode a successful frame exchange should happen (e.g. Data with PM=1, and receiving ACK), however in your proposal you are setting the power save and sleep mode in beacon or probe response without acknowledgment, so it may not be received by some of the STAs of non-AP MLD due to different reasons. In this case, AP MLD goes to doze state while STAs are not notified of this change, so they stay awake and keep sending the UL frames.

[Guogang Huang]Correct! The current Spec. only allows the STA to enter the power save mode. And The STA will only communicate with the associated AP. Hence after a successful frame exchange, it can enter the power save mode with the doze state. But if we consider the power save of the AP MLD, since it serves many STAs, the above method cannot be used. To address your question, the AP MLD shall advertise the power management info in advance. The related text is shown as below.

[Morteza] I think advertising the PM info in advance doesn't necessarily results in successful notification of PM change. How about using the triggering mechanism to poll the multiple STAs for PM change?

* If you define a Sleep mode in addition to power save and active mode, then non-AP MLDs can have such a mode too in which it doesn't need to wake up for the DTIM beacon to save more power; in this case, AP MLD sends it's PM information only over the other link through RNR.

[Guogang Huang] The reason why I define a sleep mode is based on some member’s comments. It is used in the following cases, e.g. AP maintenance, internal interference avoidance and so on. In these use cases, the AP cannot be waked up. For the power save of the non-AP MLD, I don’t find why we need to define a sleep mode for the non-AP MLD.

[Morteza] I mentioned the use case of Sleep mode for STA in my above comment; when the AP is in Sleep mode, the STA doesn't need to wake up for the DTIM beacon (there is no beaconing over that link) and can save more power, so STA can go to Sleep mode too.

* What is the behavior if the AP of the AP-MLD is in power save mode but STA of the non-AP MLD operating on the same link switches to active mode (lets say AP sends the trigger frame after receiving the AAR control and STA sends the response with changing PM=0)? I think this is not defined in your doc.

[Gougang Huang]If the AP is operating in the power save mode with the doze state, I don’t see any benefit to let the STA being in the active mode or the power save mode with the awake state. Under this case, the STA should be in the power save mode with the doze state. If some STA wants to use this link for the delivery, it shall send a wakeup request through another link. When the AP operating in the power save mode is waked up, it shall initialize the power management mode as the power save mode with the doze state for the STAs who didn’t send a wakeup request. But for the STA who previously send a wakeup request, the AP shall regard it is operating in the active mode or the power save mode with the awake state, which depends on the setting of the Power Management bit.

[Morteza] I was referring to the case where AP is in PS and awake state, not the doze state.

So you want to prohibit the STA of non-AP MLD to go to Active mode when the AP is in PS mode over that link. I think this is missing in your proposal.

Regarding the STA in PS mode when the AP is in PS mode: STA can be in PS mode and awake state to poll the AP over the same link by sending PS-Poll for example and if the AP responds with ACK, STA can start it's UL transmission (AP is in awake state (in PS mode) with higher probability as it's responding to multiple STAs, so the response to PS-Poll can be successful with a good chance); something to think of it more.

Thanks,

Morteza

From: Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>
Sent: Thursday, March 31, 2022 4:16 PM
To: huangguogang <huangguogang1@xxxxxxxxxx>; 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx>; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
Subject: RE: Discussion on power save of AP MLD

Hi Guangguo,

I’m asking the unassociated legacy STA connection issue, you answer on the non-AP MLD staff…

Hi Guogang, Gaurang, Arik,

I see all of you have different proposals to indicate the AP unavaible state for all kinds of reason, I believe each scenario is important during the online/offline talk.

I’m considering whether we can have a uniform solution to cover so many different scenarios, or do we really need one proposal match one case? For me, I prefer to former one.

Thanks

Best Regards

Jay Yang

From: huangguogang <huangguogang1@xxxxxxxxxx>
Sent: 2022年3月31日 15:16
To: Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
Subject: 答复: Discussion on power save of AP MLD

Hi Jay,

Thanks for your question. Please find my response inline.

Regards

Guogang Huang

发件人: Yang, Zhijie (NSB - CN/Shanghai) [mailto:zhijie.yang@xxxxxxxxxxxxxxx]
发送时间: 2022年3月31日 15:09
收件人: huangguogang <huangguogang1@xxxxxxxxxx>; 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
抄送: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
主题: RE: Discussion on power save of AP MLD

Hi Guoguang,

l the AP operating in the power save mode may need periodically to wake up and transmit the Beacon frame

Please note such design will cause the legacy STA connection failure issue if it intends to set up connection based on the received Beacon.

That’s why 802.11 specification has the AP always powering on.

[Guogang Huang]For this question, I add the following note to explain it in my contribution. To optimize the multi-link performance, especially in the enterprise scenario, I think it is useful to manage the non-MLD STAs and only allow them to associate in specified links.

Thanks

Best Regards

Jay Yang

From: huangguogang <huangguogang1@xxxxxxxxxx>
Sent: 2022年3月31日 14:57
To: 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
Subject: 答复: Discussion on power save of AP MLD

Hi Xiangxin,

Thanks for your questions. Please find my response inline.

Regards

Guogang Huang

发件人: 顾祥新 (Xiangxin Gu) [mailto:Xiangxin.Gu@xxxxxxxxxx]
发送时间: 2022年3月31日 13:35
收件人: huangguogang <huangguogang1@xxxxxxxxxx>; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
抄送: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
主题: RE: Discussion on power save of AP MLD

Hi Guogang,

Thanks for the interesting proposal!

I have questions to expand the mechanism to AP MLD besides mobile AP MLD:

How to evaluate signal condition of the link that the non-AP MLD is going to request the AP MLD to wake up.

[Guogang Huang]Good question. Maybe we can consider the following approaches to address this issue:

l the AP operating in the power save mode may need periodically to wake up and transmit the Beacon frame

l Or the non-AP MLD may decide whether it sends a wakeup request according to some kind of the link quality estimation. When the AP in the power save mode transitions from the doze state to the awake state, then the AP can transmit the Beacon frame or do the channel sounding on that link.

For an AP affiliated with an AP MLD in doze, what is the TBTT Offset carried RNR in the Beacon transmitted by another AP affiliated with the same AP MLD?

[Guogang Huang]In this case, I suggest when a reported AP is operating in the power save mode or the sleep mode, only the MLD parameters subfield is carried. As I mention before, if we allow the AP in the power save mode to periodically wake up to send the Beacon frame, then your question can be resolved. For this part, I’m open to discuss.

BR,

Xiangxin Gu

Communication Standards Devision

From: huangguogang <huangguogang1@xxxxxxxxxx>
Sent: Thursday, March 31, 2022 10:03 AM
To: Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Jarkko Kneckt <jkneckt@xxxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; 顾祥新 (Xiangxin Gu) <Xiangxin.Gu@xxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
Subject: 答复: Discussion on power save of AP MLD

Hi Kaiying,

Thanks for your discussion. Please find my response inline.

Regards

Guogang Huang

发件人: Kaiying Lu [mailto:Kaiying.Lu@xxxxxxxxxxxx]
发送时间: 2022年3月31日 4:01
收件人: Jarkko Kneckt <jkneckt@xxxxxxxxx>; huangguogang <huangguogang1@xxxxxxxxxx>
抄送: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; "顾祥新 (Xiangxin Gu)" <Xiangxin.Gu@xxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
主题: RE: Discussion on power save of AP MLD

Hi Guogang,

Thanks for the contribution.

I have two general comments for clarification first.

Regarding to “power save mode (10)”, “In this mode, the AP is allowed to transition between the awake state and the doze state. The AP operates in the doze state by default unless it receives a wakeup request and enters the awake state.”

Do you mean that the AP is not allowed to transition between doze and awake itself?

[Guogang Huang]Maybe I get your point. You mean if the AP MLD has DL buffered BU, it also can transition from the doze state to the awake state, right? I agree with you and will fix it in the next revision.

Regarding to “sleep mode (11)”, “In this mode, the corresponding AP is always in the doze state. And The non-AP MLD cannot send a wakeup request to this AP through its affiliated STA.”

What do you mean by “AP is always in the doze state”? Can the AP be awake at some point? Otherwise, the AP is just as disabled.

[Guogang Huang]This mode is used in the following use cases, e.g. the AP maintenance and regulatory reasons. During this period, I think the AP cannot be ready. But when the maintenance is finished, the AP can switch from the sleep mode to the active mode or the power save mode.

Please correct me if I misunderstood something.

Thanks,

Kaiying

From: Jarkko Kneckt <jkneckt@xxxxxxxxx>
Sent: Wednesday, March 30, 2022 9:12 AM
To: huangguogang <huangguogang1@xxxxxxxxxx>
Cc: STDS-802-11-TGBE@xxxxxxxxxxxxxxxxx; Kaiying Lu <Kaiying.Lu@xxxxxxxxxxxx>; Chunyu Hu <chunyuhu@xxxxxx>; Park, Minyoung <Minyoung.park@xxxxxxxxx>; Morteza Mehrnoush <mmehrnoush@xxxxxx>; Laurent Cariou <laurent.cariou@xxxxxxxxx>; Yang, Zhijie (NSB - CN/Shanghai) <zhijie.yang@xxxxxxxxxxxxxxx>; Yonggang Fang <Yonggang.Fang@xxxxxxxxxxxx>; "顾祥新 (Xiangxin Gu)" <Xiangxin.Gu@xxxxxxxxxx>; Rubayet Shafin <r.shafin@xxxxxxxxxxx>; Rojan Chitrakar <rojan.chitrakar@xxxxxxxxxxxxxxxx>; Liwen Chu <liwen.chu@xxxxxxx>; Abhishek Patil <appatil@xxxxxxxxxxxxxxxx>; Gaurang Naik <gnaik@xxxxxxxxxxxxxxxx>; Srinivas Kandala <srini.k1@xxxxxxxxxxx>; Xiaofei Wang <Xiaofei.Wang@xxxxxxxxxxxxxxxx>; NEZOU Patrice <Patrice.Nezou@xxxxxxxxxxxx>; Ganming(Ming Gan) <ming.gan@xxxxxxxxxx>
Subject: Re: Discussion on power save of AP MLD

Hi Guogang, thank you for your contribution.

I have two comments:

1. 802.11be has disabled link concept that (no TIDs mapped to a link) could be used to enable long term affiliated AP sleep.

- I am wondering could the AP sleep mode be implemented by using disabled links?

2. Typically regular APs have not operated in power save, but have supported aggressive power save of the STAs.

- Have you considered to make power save enhancements (mode 10) to be available only for the NSTR Mobile APs?

- I think mobile APs could have more aggressive power save mechanisms.

Cheers,

Jarkko

On Mar 30, 2022, at 09:00, huangguogang <huangguogang1@xxxxxxxxxx> wrote:

Hi all,

I have presented my contribution 22/356r5 in the call. But there is no time to take questions. I initiate this email thread to further discussion for doc 22/356. Please let me know your questions and comments.

Regards

Guogang Huang

This email (including its attachments) is intended only for the person or entity to which it is addressed and may contain information that is privileged, confidential or otherwise protected from disclosure. Unauthorized use, dissemination, distribution or copying of this email or the information herein or taking any action in reliance on the contents of this email or the information herein, by anyone other than the intended recipient, or an employee or agent responsible for delivering the message to the intended recipient, is strictly prohibited. If you are not the intended recipient, please do not read, copy, use or disclose any part of this e-mail to others. Please notify the sender immediately and permanently delete this e-mail and any attachments if you received it in error. Internet communications cannot be guaranteed to be timely, secure, error-free or virus-free. The sender does not accept liability for any errors or omissions.
本邮件及其附件具有保密性质，受法律保护不得泄露，仅发送给本邮件所指特定收件人。严禁非经授权使用、宣传、发布或复制本邮件或其内容。若非该特定收件人，请勿阅读、复制、使用或披露本邮件的任何内容。若误收本邮件，请从系统中永久性删除本邮件及所有附件，并以回复邮件的方式即刻告知发件人。无法保证互联网通信及时、安全、无误或防毒。发件人对任何错漏均不承担责任。

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