| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Hi Giovanni,
Thank you for your comments.
I agree that an additional delay may be introduced if the UL transmitted to the current AP MLD need to be forwarded to the target AP MLD.
However, the main motivation is not only to reduce the UL OTA delay component, but also to avoid packet discard during ST execution.
If UL transmission is not allowed during ST execution, some packets may not be transmitted at all and may eventually be discarded due to lifetime expiration.
Therefore, even if forwarding introduces some additional delay, allowing UL transmission during ST execution may still be beneficial compared with a case where the packets cannot be transmitted at all.
Regarding your second point, I understand the concern. I am not fully sure whether the non-AP MLD is actually unable to recover the MSDUs from the prepared MPDUs and repackage them for the target AP MLD. However, if such a risk exists, then direct UL transmission
to the target AP MLD during ST execution could be another possible approach.
So, I think there are two possible approaches to reduce UL OTA delay and UL discard during ST execution:
- transmission to the current AP MLD, as introduced in 26/0396; and
- direct transmission to the target AP MLD during ST execution.
The first approach may be useful when transmission to the current AP MLD is still feasible and forwarding to the target AP MLD can be handled by the infrastructure.
The second approach may be preferable if re-packaging MPDUs at the non-AP MLD is challenging, or if transmission to the current AP MLD becomes unreliable.
Based on this, I think the additional spec text may still be worth considering.
Best regards,
Ryuichi
差出人: Giovanni Chisci <gchisci@xxxxxxxxxxxxxxxx>
送信日時: 2026年6月5日 4:38 宛先: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx>; Hirata, Ryuichi (SEC) <Ryuichi.Hirata@xxxxxxxx> 件名: Re: Discussion on 11-26/0396(UL transmission in SMD BSS Transition)
Dear Ryuichi,
Thanks for preparing this contribution.
Here are my thoughts:
- the goal of the spec text developed in 26/0396 seems to be to allow non-AP MLD to transmit UL to current AP MLD after the ST execution is started, to allow getting this UL out OTA, to perhaps reduce the OTA delay component. I think that the benefits of this
on the overall of UL are unclear, since the additional delay will be re-inserted due to the need of forwarding to that target AP MLD.
- risk: When the client prepares an MPDU for the current AP MLD with MSDUs, it is unclear whether the non-AP MLD would be able to recover such MSDUs and repackage them into a new MPDU for the target AP MLD, for example in the case of the link with the current
AP MLD being unreliable/dying. So there is a tangible risk to lose even more MSDUs if the client keeps insisting to package and transmit more MPDUs for the current AP MLD.
Happy to hear your thought on the above, I am currently unsure that the additional spec/implementation complexity is warranted for adoption.
Best,
Giovanni
From: Ryuichi.Hirata@xxxxxxxx <Ryuichi.Hirata@xxxxxxxx>
Sent: Thursday, June 4, 2026 08:59 To: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx> Subject: [STDS-802-11-TGBN] Discussion on 11-26/0396(UL transmission in SMD BSS Transition) WARNING: This email originated from outside of Qualcomm. Please be wary of any links or attachments, and do not enable macros.
Hi all,
I am sending this email to kick off the offline discussion on document 26/396, which I presented during the joint call.
Please feel free to share your comments or questions.
Best regards,
Ryuichi
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 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 |