Hi everyone,
I received questions on overhead and complexity and the Retry bit. Here’s my answer.
Mark provides another solution with no overhead either. See it at the end.
Questions and comments and views are welcome.
Thanks!
Overhead can be seen by comparing 2 MPDUs as following.
A-Control expansion with A-Control ID ONES (4 Octets is inserted after HT Control field).
A-Control expansion proposed by document 23/2015r2 (A-Control Expansion field overrides HT Control field)
Complexity can also be seen by comparing them.
With
A-Control expansion with A-Control ID ONES, we need following steps:
HW extracts HT Control field.
HW parses HT Control field.
If (the HT Control field is HE Variant) {
HW extracts A-Control subfield.
HW parses A-Control subfield (partially).
If (the A-Control subfield is A-Control ID ONES) {
HW extracts the new added A-Control Expansion field.
HW/SW parses the new added A-Control Expansion field.
}
HW extracts Data field.
}
else {
HW extracts Data field.
}
With
A-Control expansion proposed by document 23/2015r2 , it is easier.
HW extracts HT Control field
HW extracts Data field
If (B11 is 0) {
HW/SW parses the HT Control field as HT Control field.
Else {
HW/SW parses HT Control field as A-Control Expansion field.
}
What will happen for frames where the Retry bit is not reserved?
For exchanged QoS Data or Management frames between UHR STAs that are non-A-MPDUs and not under a block ack agreement, the receiver side always search the appropriate RC to see whether it is
a retransmission or not. So no mistake will be introduced. It is worth to point out that these frames are few percent.
Actually, Mark provides another solution with no overhead either.
UHR STAs can have 26-bit A-Control Expansion subfield after A-Control ID ONES
|
Control ID value |
Meaning |
Length of the Control Information subfield (bits) |
Content of the Control Information subfield |
|
…… |
|
|
|
|
15 |
Ones need expansion surely(ONES) |
26 |
For UHR STAs: A-Control Expansion subfield For non-UHR STAs:
set to all 1s |
I think this solution brings the constraint that only 23 or 22 bits for delivering information, taking aside 3 or 4 bits for next level Control ID. And it also leads to complex processing.
BR,
Xiangxin Gu
Communication Standards Devision
From:
顾祥新 (Xiangxin Gu)
Sent: Thursday, August 15, 2024 9:23 AM
To: STDS-802-11-TGBN@xxxxxxxxxxxxxxxxx
Subject: Any questions or concerns on document 23/2015r2?
Dear all,
I presented document 23/2015r2, which provides an alternative scheme for A-Control expansion for transmission between 11bn STAs as following.
This scheme has lower overhead and complexity, comparing to using the reserved A-Control ID ONES.
Document
23/2015r2 provides more details.
If you have questions or concerns, please feel free to contact with me.
Thanks!
BR,
Xiangxin Gu
Communication Standards Devision
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