Re: [STDS-802-11-TGBN] Seamless roaming PDT Part 3, Part 4, and Part 5

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

Hi Xuwen,

Thank you for your questions and please see my responses inline below.

From: Zhao Xuwen <zhaoxuwen123@xxxxxxxxxxx>
Sent: Tuesday, July 22, 2025 12:30 AM
To: Duncan Ho <dho@xxxxxxxxxxxxxxxx>; STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx
Subject: 回复: [STDS-802-11-TGBN] Seamless roaming PDT Part 3, Part 4, and Part 5

WARNING: This email originated from outside of Qualcomm. Please be wary of any links or attachments, and do not enable macros.

Dear Duncan：

Thank you for preparing these PDTs. Please find my comments as below:

In PDT part3:

Regarding comments on the security context, I believe it is not appropriate to simply reject relevant comments based on implementation reasons. At the very least, discussions on the security context should be grounded in the existing conclusions of the security mechanisms (which can be supplemented in section 37.9.8 "Context"). Taking "Different PTK mode" as an example, according to the current description in section 37.9.5.3 "Different PTK Key Derivation," the non-AP MLD and target AP MLD need to derive the same new PTK based on the same PMK, ANonce, and SNonce information. Therefore, there must be an interaction involving PMK ID, ANonce, and SNonce between the two parties. These details should be explicitly written into the standard to prevent failures caused by differences in implementation approaches between vendors (e.g., between the current AP MLD and the target AP MLD) during roaming.

[DH] We can discuss again. I was just following the direction most have been assuming in the past few years during the development of ST.

Regarding PTK Key Derivation in the Different PTK mode, I believe the current derivation method described in the text is relatively complex. The original intention of seamless roaming is to minimize latency. For centralized PTK derivation, it may be worth considering reusing previous information as much as possible to save time while ensuring security. For example, as I suggested in CID 273: the non-AP MLD and the current AP MLD generate a PTK identifier while generating the old PTK before roaming. The target AP MLD obtains the old PTK from the current AP MLD based on the PTK identifier. The non-AP MLD and the target AP MLD then derive a new PTK based on the old PTK.

[DH] The text I used follows a lot of offline discussions and it’s my best estimate of the highest chance of convergence at this point in time. I would encourage you to discuss your proposal with other interested members and try to converge.

Thanks,

Duncan

Thanks,

Duncan

B.R.

Xuwen Zhao

TCL

发件人: Duncan Ho <00002b3e54cff3e2-dmarc-request@xxxxxxxxxxxxxxxxx>
发送时间: 2025年7月19日 13:23
收件人: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx
主题: [STDS-802-11-TGBN] Seamless roaming PDT Part 3, Part 4, and Part 5

Dear Seamless Roaming TTT,

I’ve prepared these documents for the ad-hoc meeting next week. Could you please let me know if you have any comments or questions?

25/753r2 PDT MAC on Seamless Roaming Part 3 (Security) – the r1 was presented before so this is an updated version

25/1020r0 PDT-CR MAC on Seamless Roaming Part 4 (Emergency roaming) – not presented before

25/1101r0 PDT-CR MAC on Seamless Roaming Part 5 (Detailed frame formats and signaling) – not presented before

Thanks,

Duncan

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