| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Hi Yelin, Pooya, Thank you for the discussion on this topic. TGbe has defined a mechanism to uniquely identify each AP MLD that have an affiliated AP in the same multiple BSSID set (i.e., AP MLD ID = index of the BSSID) and for any other AP MLD (i.e., non-collocated), the AP MLD ID is >= 2^n (n=MaxBSSID
Indicator). With that view in mind, when an AP affiliated with an AP MLD advertises another AP MLD which has AP MLD ID >= 2^n (n=MaxBSSID Indicator) the non-AP MLD can determine if the reported AP MLD is collocated or non-collocated. In addition, the Collocated
bit in the reported AP (MLD) in the RNR IE also helps identify collocated or non-collocated-ness (this is useful for the case of co-hosted BSSID set or when the affiliated AP is standalone and of course it still applies to MBSSID set). Therefore, the necessary
signaling is already provided by TGbe and will continue to apply in TGbn (i.e., no new work needed). In addition, the hierarchy defined by TGbe will continue to hold in TGbn – i.e., AP MLDs within a multiple BSSID or co-hosted BSSID set will be affiliated
with different SMD MLDs (and are connected to different ESS/DS and advertise a different SSID). Regards, From: Pooya Monajemi <00000ef0b9e0aff7-dmarc-request@xxxxxxxxxxxxxxxxx>
WARNING:
This email originated from outside of Qualcomm. Please be wary of any links or attachments, and do not enable macros. Many thanks for the response Yelin, I agree that there needs to be a method for identifying neighboring AP MLD's that are not co-located but are part of the same logical entity above (UFT in your contribution).
We should be able to assign a MAC address to this logical entity and announce it to achieve the identification problem. Also open to hearing other opinions. Best Pooya On Thursday, January 18, 2024 at 12:00:25 PM EST, Yelin YOON <yl.yoon@xxxxxxx> wrote:
Hello Pooya. To give a brief history on how we came up with the Collocated AP MLD Set, in roaming, we consider a case where the non-AP STA roams
between the AP MLDs that are not co-located. Therefore, a non-AP STA should be able to identify the AP MLDs that are not co-located and provided by the serving AP MLD.. However, as I mentioned in my contribution (1907), there is no method to identify the AP MLDs that are not collocated.
While thinking of the method on the identification of the non-collocated AP MLDs, we also wanted to take the multiple BSSID set case
into the consideration as well where the APs affiliated with different AP MLDs included in the same multiple BSSID set.
This is the reason that we created a co-located AP MLD set that groups the AP MLDs that are co-located and set an ID to overcome the
identification limitation. On the other hand, we do not have a firm opinion on creating the co-located AP MLD set and we are open to roaming between AP MLDs individually.
If any of the other members have their thoughts on this, I would be welcomed to hear from you. Best Regards,
Yelin Yoon
From: Pooya Monajemi [mailto:pmonajemy@xxxxxxxxx]
Hello Yelin and thanks again for the contribution. On the question of why we need the definition of a co-located AP MLD set, you mentioned cases where two AP MLDs are co-located and STA is
roaming from one to another. I believe today if two co-located APs are advertising the same SSID (~same ESS) they generally belong to the same AP MLD, do you see the
case of same SSID on different but co-located AP MLDs today or in future? Thanks Pooya 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 |