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

[STDS-802-11-TGBF] 答复: [STDS-802-11-TGBF] 答���: Follow-up on Local Reporting for 11bf ..

Hi Oscar,

 

Thanks for your detailed explanations. According to your feedback, can I understand your thoughts in this way?

1.     You want to define that in some cases, there is no need to give the initiator the sensing results, which is different from the current SFD. For example, in the distributed portion, the receiver does not need to feed back their results.

2.     What about this case: The AP is the initiator, and the receiver STAs do the locally reporting and calculation. Then those STAs can get some sensing results locally (such as whether there is a motion in the room). If these detection results are fed back to the AP through applications or MAC/PHY later, can we call this an example of the Combo 3 you listed below?

3.     You want to support all these three cases you listed here (local, nonlocal, and hybrid).

Thanks.

 

 

Best regards,

Mengshi Hu

 

 

发件人: Oscar Au <oscar.au@xxxxxxxxxxxxxxxxxxxx>
发送时间: 2021年12月8日 5:51
收件人: humengshi <humengshi@xxxxxxxxxx>
抄送: STDS-802-11-TGBF@xxxxxxxxxxxxxxxxx
主题: Re: [STDS-802-11-TGBF] 答复: Follow-up on Local Reporting for 11bf ..

 

Dear Mengshi,

 

Thank you for your question. My answers are inline.

 

On Mon, Dec 6, 2021 at 6:52 PM humengshi <0000179fd8c5ace5-dmarc-request@xxxxxxxxxxxxxxxxx> wrote:

Hi Oscar,

 

Thanks for bring these contributions. I have one question:

Q1: In Slide 7 in 1909, you give two examples on the use of local reporting. Regarding the second one, when the AP is the initiator, you are saying that the measurements can be done locally and no transmission of measurement results over the air. I’m a little bit confused about this. Since the AP is the initiator, the AP shall know the final results. Here you mention that no results are transmitted. How does the AP get the sensing results?

Good question. 

 

We can roughly divide a WLAN sensing task (e.g. motion detection) into a distributed portion (e.g. motion detection for one of the rooms in a house) and a centralized portion (e.g. motion detection for the whole house). 

- The distributed portion uses multiple STAs to (a) generate CSI (i.e. sensing measurement results) locally, and (b) use the locally-generated CSI to perform local (or distributed) WLAN sensing computing. There may be one STA (e.g. smart speaker, TV, refrigerator, lamp, camera, etc) in each room in the house.

- The centralized portion (a) concentrates all CSI from the multiple STAs (transmitted over the air) at one STA (sensing initiator), and (b) uses the global CSI (i.e. all the CSI) to perform global (or centralized) WLAN sensing computing. The initiator may be an AP or a hub in the house.

 

What I mean in the second example of Slide 7 of 1909 is that no transmission of measurement results over the air is needed for the distributed portion of WLAN sensing. 

- Of course, based on our current SFD, the CSI will be reported back to the AP (the initiator), and that would imply te transmission of CSI over the air ("Nonlocal Reporting") - for the centralized portion of WLAN sensing.

 

- In 1909, I was describing several combinations (or "methods") of Nonlocal Reporting (reporting based on SFD) and Local Reporting.

- Combo 1 is Nonlocal Reporting only. In this case, the CSI are sent to AP for the centralized portion. The distributed portion is not supported.

- Combo 2 is Local Reporting only. In this case, the CSI are locally generated/reported for the distributed portion. Centralized portion is not supported.

- Combo 3 is both Local and Nonlocal Reporting. Both distributed and centralized portions can be supported.

 

I hope these clarify the situations.

 

Thanks,

Oscar

 

 

Best regards,

Mengshi Hu

 

发件人: stds-802-11-tgbf@xxxxxxxxxxxxxxxxx <stds-802-11-tgbf@xxxxxxxxxxxxxxxxx> 代表 Oscar Au
发送时间: 2021年12月7日 9:34
收件人: STDS-802-11-TGBF@xxxxxxxxxxxxxxxxx; Hanxiao (Tony, WT Lab) <tony.hanxiao@xxxxxxxxxx>
抄送: Oscar Au <oscar.au@xxxxxxxxxxxxxxxxxxxx>
主题: Follow-up on Local Reporting for 11bf ..

 

Dear all,

 

Thank you for the many valuable comments on the topic of Local Reporting for 11bf (1904, 1905, 1906, 1908, 1909). I understand that many colleagues have questions on Local Reporting. With this email thread, I would like to invite you to share comments and thoughts on Local Reporting.

 

To summarize what "Local Reporting" is:
1. Local reporting is to make sensing measurement results (e.g. CSI) locally available in the sensing receiver to upper layers (e.g. application layer) in the  sensing receiver through some MLME primitive, with no requirement of transmission of the sensing measurement results.

 

2. Use cases. In 1909, two "recipes" (examples) are shown on the use of local reporting to achieve central-computing and distributed-computing WLAN sensing solutions. What do you think about those two recipes? What are your concerns?

 

3. Further details. There are some comments that more details or more understanding of local reporting is needed. Please kindly share what details we should try to investigate/explore. What are your concerns?

 

Thanks,

Oscar

 

 

 

 

To unsubscribe from the STDS-802-11-TGBF list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBF&A=1

To unsubscribe from the STDS-802-11-TGBF list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBF&A=1

To unsubscribe from the STDS-802-11-TGBF list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBF&A=1