| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Hi Sindhu,
Thank you for your comments. I really appreciate your feedback.
Regarding the first two paragraphs, I see why you might consider the behavior obvious. However, since we received at least 10 CIDs asking for normative clarification on this issue, I believe we need to explicitly specify the behavior without over-constraining
implementations.
From a protocol state machine perspective, D1.5 states that the Co-BF Trigger frame is transmitted "after the end of the PPDU carrying the ICR." If an ICR is not received, this condition is technically never satisfied. Introducing explicit normative text (like
"shall not") is necessary to resolve this ambiguity, prevent potential deadlocks, and guarantee interoperability.
Nevertheless, I agree that we must avoid hindering the AP's implementation flexibility. If you have any concerns regarding the behavior itself, please let me know.
Finally, I completely agree with your comment on the last paragraph. I would like to consider this a bit further, so I am planning to mark the related comments as "Deferred" for the next SP.
Best regards,
Kosuke (SONY)
差出人: Sindhu Verma
送信: 2026 年 6 月 25 日 (木曜日) 18:23 宛先: Aio, Kosuke (SEC) Cc: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx 件名: On "26/0314 LB291 CR for Co-BF and Co-SR Error Recovery Sequence" Hi Kosuke,
Your first 2 proposed paragraphs are trying to mandate that if none of the EMLSR/DPS non-APs of an AP respond with ICR and there were no other non-APs in the corresponding CBF-invite/response, the AP does not proceed with the CBF sequence. I think this is obvious behavior and any variations can be left to the AP implementation depending on its reaction time. If we do not trust that, we need to specify additional obvious behavior such as the coordinated AP not transmitting an MU-BAR if it does not transmit a CBF PPDU. Doesn’t the current CBF sequence already state that the frames must be exchanged in that particular order?
Your 3rd proposed paragraph ((#7382) When the Co-BF coordinating AP schedules to transmit another Co-BF Trigger frame within the current TXOP, the Co-BF coordinated AP may transmit a CF-End frame to the Co-BF coordinating AP in place of an MU-BAR Trigger frame to indicate the termination of its participation in the Co-BF transmission sequence if the Co-BF co-ordinated AP cannot participate in the previous transmission. When receiving the CF-End frame from the Co-BF coordinated AP, the Co-BF coordinating AP shall not transmit another Co-BF Trigger frame to the Co-BF coordinated AP until a new Co-BF Invite and Co-BF Response frame exchange is completed.) alters the currently specified CBF data sequence and conflicts with what has already been agreed in D1.5 on TXOP termination: (#6530, 11447)The coordinating AP should transmit a CF-End frame at the end of the sequence to terminate it. The CF-End frame shall be transmitted by the coordinating AP 2×aSIFSTime plus the duration of the acknowledgment sequence between the Co-BF coordinated AP and its scheduled non-AP STAs indicated by the Ack Sequence Duration field in the Co-BF Response frame, after the end of the PPDU carrying the BlockAck or Multi-STA BlockAck frame(s) transmitted by the recipient non-AP STA(s) associated with the Co-BF coordinating AP. The frame sequence is shown in Figure 37-10 (The Co-BF transmission sequence including CF-End frames transmitted by the coordinating AP and optionally by the coordinated AP).
(#6530, 11447)The coordinated AP should transmit another CF-End frame aSIFSTime after the end of the first CF-End frame transmitted by the coordinating AP. The coordinated AP shall not transmit a CF-End frame unless the coordinating AP has already transmitted a CF-End frame. The CF-End frame transmission by the coordinated AP is also shown in Figure 37-10 (The Co-BF transmission sequence including CF-End frames transmitted by the coordinating AP and optionally by the coordinated AP). (#6530)The non-AP STA(s) associated with the coordinating AP or the coordinated AP, and that is applying the extended timeout period may interrupt that extended timeout period and switch to either LC mode for DPS STAs or listen mode for EMLSR STAs once they receive any of the CF-End frames. (#6530) NOTE 3—The CF-End frame serves as a Co-BF TXOP termination indication to the coordinated AP as well as the non-AP STAs associated to the coordinating and the coordinated APs.
In view of the above, I request that you withdraw your 3rd paragraph and modify the first two to possibly state expected behavior in the form of a note.
Regards, Sindhu
To unsubscribe from the STDS-802-11-TGBN list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBN&A=1 |