Re: [STDS-802-11-TGBN] 答复: [STDS-802-11-TGBN] Seamless roaming PoC discu

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

Hi Pooya , All,

Sorry for my delay response.

For roaming subtopics, I'm OK to volunteer the POC of UL data plane subtopics, and glad to work together on the whole data plane enhancement.

Thanks

Best Regards

Jay Yang (杨志杰)

Original

From: PooyaMonajemi <00000ef0b9e0aff7-dmarc-request@xxxxxxxxxxxxxxxxx>

To: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx>;

Date: 2024年10月19日 06:52

Subject: Re: [STDS-802-11-TGBN] 答复: [STDS-802-11-TGBN] Seamless roaming PoC discussion

Hello all,

Good that this is being discussed this in the wider forum.

As a summary we have the following list of PoC volunteers:

Duncan, Mike, Yelin , Giovani , Binita , Po-Kai, Guogang , Jay, Pooya , Jarkko , Liwen

A number of topics has been proposed but was not discussed much, at least not with the TTT . The PoC should check that topics proceed and that quality of the contributions. Many of the topics require large contributions that are not easy to split between very narrowly split PoCs .

Actually the list is also missing one item

We propose the following 6 PoCs :

- Architecture and Security

- AP Discovery

- Roaming assurance/preparation

- Roaming signaling

- FT enhancements

- Data plane

Here is short justification for the modifications to the previous proposals:

- Architecture and Security framework should be combined to the same topics as Guogang proposed. The security is the key aspect of the architecture and we need to ensure that both aspects are taken case.

- Roaming assurance/preparation: This group should focus on ensuring successful roaming (roaming assurance) and preparing the roaming to minimize the actual roaming time.

- Roaming signaling should cover the STA-AP signaling and the context transfer signaling, because they are likely very likely related to each other.

- Data plane: For the Data plane topic, maybe Jay and I can work on the data plane together. UL/DL split is a bit too much division imho , and also I have a number of topics for DL handling (or Jay please let me know if you'd be ok with doing UL instead)

Best

Pooya and Jarkko

On Thursday, October 17, 2024 at 10:10:20 PM PDT, Binita Gupta <bingupta.ieee@xxxxxxxxx> wrote:

Hi Duncan, Guogang , all,

Thanks for this discussion.

I see architecture and security as separate subtopics. Specifically, the Architecture elements would be captured in clause 4 and 5 (and possibly in new UHR MAC clause) as part of Reference models, data plane architecture etc, as was done for MLD . The security aspects will be captured as part of security clause 12 and new UHR clause. So, I don't think they need to be under one PoC as Guogang is suggesting.

Also, I don't agree with the division of subtopics suggested by Guogang . Specifically, the "AP-assisted roaming (e.g. recommend)" is same as Discovery of candidate APs for roaming. So there is no reason to have two separate subtopics for the same functionality. Also, I have made contributions on preparation and roaming execution, so I volunteer to be PoC for those subtopics. Furthermore, I think discovery, preparation, and roaming execution would define the roaming signaling details for those subtopics, so I don't think we should have a separate PoC for roaming signaling. Signaling aspects will be defined as part of each subtopic by PoCs , and would be the key part of the PDTs produced.

I think splitting the DL and UL data handling is fine.

So, I propose to keep Duncan's suggestion with potentially that change:

Architecture/framework - Duncan

Security - Mike

Discovery – Yelin

Link Preparation – Giovanni/Binita

Roaming execution – Binita /Giovanni

Context transfer - Po-Kai

FT enhancements – Guogang

DL Data plane – Jay

UL Data plane – Pooya

Thanks,

Binita

On Thu, Oct 17, 2024 at 8:41 PM huangguogang <000017b1384624cd-dmarc-request@xxxxxxxxxxxxxxxxx> wrote:

Hi Duncan and all,

Thanks for this discussion. I would like to express my opinions as follows.

Firstly, I agree with Jarkko ’s comment, the roaming architecture and security are highly coupled with each other. There is no any reason to separate them. For this sub-topic, I think we should select a security expert as the PoC . Mike should be the best suitable person in my mind.

Secondly, the sub-topic division proposed by Duncan is unclear to me. Nobody knows the differences of these sub-topics.

Based on the above comments and contributions of each candidate, I propose the following scheme for discussion. Except the first subtopic, I’m open to discuss for other sub-topics/changes.

Architecture & security – Mike
AP-assisted roaming (e.g. recommend) – Binita
Discovery – Yelin
Link Setup – Duncan
Context Transfer – Po-kai
Data Forwarding – Jay
Retrieval Buffered Packet or data plane– Pooya
Roaming Signaling (Clause 9) – Guogang Huang

Regards
Guogang Huang
发件人: Binita Gupta <bingupta.ieee@xxxxxxxxx>
发送时间: 2024年10月18日 9:50
收件人: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx
主题: Re: [STDS-802-11-TGBN] Seamless roaming PoC discussion

Hi Duncan, Jarkko , all,

I think roaming subtopics as suggested by Duncan seems very logical and reasonable partition into subtopics.

I think splitting architecture/framework and security as two subtopics does makes sense. The architecture would cover the overall roaming architecture & framework and security is a separate subtopic that would focus mainly on security aspects of roaming. Sure, there are some related aspects and PoCs +TTTs would collaborate to align on concepts/directions etc following the TGbn guidelines laid out in the guidelines doc 24/1682.

On Discovery and Preparation - these are two different steps as proposed by many earlier roaming presentations. Discovery would select one or more candidate APs and then preparation is to prepare the selected AP(s) in advance, to later execute the roaming. Also, to avoid confusion, since the 'Roaming' subtopic refers to final roaming execution, I suggest renaming it to 'Roaming execution'.

Overall, I concur with Duncan that roaming is a big topic and to make faster progress it does make sense to divide up the subtopics among PoCs , to get things done in parallel. The PDTs that are produced by POCs+TTTs will be shared with this entire TTT group for review and commenting, so all TTT members are able to contribute and influence the PDTs .

I look forward to collaborating with you all on roaming!

Thanks,
Binita

On Thu, Oct 17, 2024 at 11:24 AM Jarkko Kneckt <00000d5619618f4f-dmarc-request@xxxxxxxxxxxxxxxxx> wrote:
Hi Duncan and all,

As discussed in the 802.11bn call this morning, the PoC assignments are open for the roaming subtopics.

I would like to volunteer for PoC of the discovery subtopic.
Pooya would like to volunteer for PoC of the data plane.

I have questions on the subgroup topics:
- Architecture and security are related topics. Would a single PoC be better? How work is split between 2 PoCs ?
- The discovery and preparation select the AP for roaming. Can you please clarify the differences of these subtopics.
- What does the roaming subtopic mean? This name is not very descriptive. All of these subtopics are related to roaming

Cheers,
Jarkko

On Oct 17, 2024, at 10:46 AM, Duncan Ho <00002b3e54cff3e2-dmarc-request@xxxxxxxxxxxxxxxxx> wrote:

Dear all,

Per the suggestion in the contribution 24/1698 (TGbn D0.1 Spec Text Volunteers and Status) from Ross, the PoC candidates should exchange email in advance trying to converge on subtopic split and PoC assignment. That’s why I started an email thread with the PoC candidates but didn’t include TTT members. I didn’t think TTT members were interested in PoC discussion since they would have signed up for PoC if they were interested in being a PoC .

Anyways, since some TTT members indicated in the 11bn meeting earlier today that they were also interested in the PoC discussion, I’ve now added all the TTT members to this thread for further discussion.

As I mentioned on the call, seamless roaming is a big topic and it could be broken down into sub-topics as multiple contributions have identified these different subtopics and phases of seamless roaming. Here’s my attempt to create the subtopics and split those amongst different PoCs . I believe Yelin , Binita , Giovanni, Po-kai and Jay are OK with the following split. I know that Guogang has some different opinions and Mike mentioned he has an upcoming contribution about architecture and requirements. Please let me know if you have any feedback.

Architecture/framework - Duncan
Security - Mike
Discovery – Yelin
Preparation – Giovanni/Binita
Roaming – Binita /Giovanni
Context transfer - Po-Kai
FT enhancements – Guogang
Data plane UL/DL delivery – Jay

I acknowledge that there are still a lot of technical discussion and agreements to be made but based on all the contributions and discussions so far, I expect what we ended up with would not deviate too much from the above subtopics. Whatever technical decision will have to pass Motion first (regardless of who the PoCs are) and that’s where the technical discussion should take place. As such, I would appreciate if we can focus on the split and subtopics and not to be derailed by any attempt of technical discussion in this thread.

BR,
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

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
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