| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Thanks Amichai. One question, given your assumption of each field, it seems that 192us UL response time will be too short.
Rojan, I believe case 2 is to introduce a new dedicated frame only for timing purpose. Case 1 is the idea to combine Ack/Nack/slot indication to a single frame, which also I prefer.
I think it would be the best to have a fixed size of the "ack/nack" frame so both AP and STA, especially STA, can calculate the (nth) slot boundary easier.
Thanks,
Sam
From: Rojan Chitrakar <00002005fb56ded7-dmarc-request@xxxxxxxxxxxxxxxxx>
Sent: Thursday, May 28, 2026 9:18 AM To: STDS-802-11-TGBP@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBP@xxxxxxxxxxxxxxxxx> Subject: Re: [STDS-802-11-TGBP] Follow-up thread on ACK indication of slot boundry Hi Amichai,
Thanks for the detailed analysis. Seems, Case 2 (Nack) is essentially an Ack with an optional RA. Two quick questions:
1. When AMP AP schedules the slots in a TXOP, what does it base its calculation of the number of slots on: Ack or Nack?
2. A non-AP AMp STA scheduled/selects a non-first slot, what does it base its calculation of the start of its slot: Ack or Nack in the previous slot(s)? Say for a 4th slot?
Different sized frames complicates the whole slot scheduling in my opinion.
Best regards, Rojan Chitrakar From:Amichai Sanderovich <00003126a116a6ee-dmarc-request@xxxxxxxxxxxxxxxxx>
To:STDS-802-11-TGBP@xxxxxxxxxxxxxxxxx <STDS-802-11-TGBP@xxxxxxxxxxxxxxxxx>
Date:2026-05-28 01:44:25
Subject:[STDS-802-11-TGBP] Follow-up thread on ACK indication of slot boundry
Hello All,
Following up on Rojan’s proposal, we would like to provide some numbers for support of the proposal.
Case 1: As we understand, it was suggested that the ACK AMP DL frame will include a TA in addition to the RA (for it to be used as start-slot indication), and that for NACK indication the RA
will be set to the broadcast address on such message.
It means that this frame can be received by all STAs, and be used by the STAs for slot boundary indication.
So the ACK will contain:
Total of the resulting ACK length is 330us for 12bits identifier and 360us for 16 bits identifier. The addition of the TA was 15% of the total frame length.
In case the solicited UL_response_time is 192us, the AP can fit 5 slots in 3ms TX-OP (deducted the trigger frame) as follows (1.01 to account for non-AP STA transmitter clock inaccuracy):
floor(3000/(SIFS+UL_response_time *1.01 + SIFS + ACK)) = 5
Case 2: The alternative is a separated dedicated new DL frame for indication of the slot.
Such frame will contain:
Total length is 281us for 12bits identifier and 297us for 16bits identifier.
The ACK in this case will be shorter by 48 or 64us for 12 or 16bits respectively.
In case the solicited UL response is 192us (as for case 1), the AP can fit 3 slots in 3ms TX-OP (deducted the trigger frame) as follows:
floor(3000/(SIFS+UL_response_time *1.01 + SIFS + ACK+SIFS + Indication_frame)) = 3
Another issue we notice for Case 2 is that a STA implementation that would like to rely on energy sensing for slot boundary can not do that since the ACK and the indication frame can only be distinguished by reception since they will have the same RSSI at the
non-AP STA.
Best Regards
Amichai
To unsubscribe from the STDS-802-11-TGBP list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBP&A=1 To unsubscribe from the STDS-802-11-TGBP list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBP&A=1 To unsubscribe from the STDS-802-11-TGBP list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGBP&A=1 |