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

Re: [STDS-802-11] TGme - PHY - discussion on RX_START_OF_Frame_OFFSET point of reference- CID 3097

--- This message came from the IEEE 802.11 Working Group Reflector ---
Hi All,
I suppose this depends on what the intended use is for the parameter. If it is intended to be used for estimation of range via time of flight measurement, the proposed definition referencing to the antenna connection does not make sense. It would ideally be at the antenna.  However as defined, 10 nsec units, the precision available would mask any difference between the actual TOA and the TOA at the antenna connector in most foreseeable uses cases (e.g. a distance of < 10' (3.048m) between the antenna and the antenna connector).   If we consider a scenario where the antenna cable may be more than 10', then it seems impossible for the implementor to know apriori what the offset value would be, which justifies the "at the connector", however this would preclude precise TOA determination.  I suppose that means this parameter is not intended to be used for precise ranging or positioning?  In which case either this shold be a R/W parameter or referenced to where the implementer can determing the value.

FWIW

Ben
From: Joseph Levy <000019588066c6b7-dmarc-request@xxxxxxxxxxxxxxxxx>
Sent: Wednesday, December 14, 2022 9:20 AM
To: STDS-802-11@xxxxxxxxxxxxxxxxx <STDS-802-11@xxxxxxxxxxxxxxxxx>
Subject: [STDS-802-11] TGme - PHY - discussion on RX_START_OF_Frame_OFFSET point of reference- CID 3097
 
--- This message came from the IEEE 802.11 Working Group Reflector ---

Dear All,

 

CID 3097 proposes to “clean-up” the definition of RX_START_OF_Frame_OFFSET.   When discussed in the TGme Ad Hoc on 12/6/22, a question was raised: should the reference point for the RX_START_OF_Frame_OFFSET be “the antenna” or “the antenna connector”?  

 

If you are a RX_STAT_OF_Frame_OFFSET expert or want to participate in this discussion, please respond to this email thread.

 

The technical point being discussed is the point of reference for the RX_START_OF_Frame_OFFSET.

  1. “antenna connector”:  The offset will be time difference between the signal arriving at the virtual or physical antenna connector point in the receive chain and the point at which the receiver time stamps the arrival.  The antenna connector is used as a point of reference for many specified requirements.  However, the location of the antenna connector in the receive chain is not required to have any defined time relationship to the WM to Antenna interface. 
  2. “antenna”: The “antenna” reference point is the WM to Antenna interface point.  (note: This may need to be clearly defined in the specification.)  The offset would be the difference in time from the arrival of the transmission at the device interface to the WM to the point at which the receiver time stamps the arrival.  

 

Please answer the question: is the point of reference “the antenna” or “the antenna connector”?  

 

The comment details and proposed change follow:

Clause 15.2.3, page 3113, line 60

Comment: 

The definition of RX_START_OF_Frame_OFFSET seems to only be located in the RXVECTOR parameter tables and the definition itself does not make sense.  All timing parameters for measurements are provided by the PHY and exist in the PHY, the MAC only uses the PHY supplied values.  Hence,   "an estimate of the offset from the point in time at which the start of the preamble off the PPDU arrived at the received antenna connector to the point in time at which this primitive is issued to the MAC" does not make sense.  In clause 8.3.5.6.3 there is a note stating that a more precise name for RX_START_OF_FRAME_OFFSET would be RX_START_OF_PPDU_TO_PHY_RXSTART_PRIMITIVE_OFFSET.  In clause 15.2.3 RXVECTOR parameters all of the parameters are defined in subclauses except the RX_START_OF_FRAME_OFFSET parameter.  However in clause 6.3.55.1 an offset is mention in Note 2 that describes an offset that can be used to correct the timestamp generated by the PHY by subtracting the delay between actual start of frame arrival at the antenna to when the timestamp is generated (this is an implementation specific value).  I believe that this offset is the RX_START_OF_FRAME_OFFSET. Also note that previous versions of 802.11 (e.g. 2012) had two times shown Figure 6-16 t2 and t3, where t3 corresponded to the time stamp and t2 the time of arrival, which was "t3"-"the offset".  Also it should be noted that the timing of when a frame is sent to the MAC has no meaning as the MAC does not measure arrival times only the PHY is capable of making such measurements.

Proposed Change (as provided by the commentor):
Replace: "An estimate of the offset (in 10 ns units) from the point in time at which the start of the preamble of the PPDU arrived at the receive antenna connector to the point in time at which this primitive is issued to the MAC." With: "An estimate of the offset (in 10 ns units) from the point in time at which the start of the preamble of the PPDU arrived at the receive antenna to the point in time that the time stamp is generated." With: "An estimate of the offset (in 10 ns units) from the point in time at which the start of the preamble of the PPDU arrived at the receive antenna to the point in time that the time stamp is generated." Note this definition should also be corrected at the following additional locations: 3114.37,  3152.6, 3168.42, 3217.6, 3235.42, 3392.27, 3523.10, and 3578.20 (9 locations in all).

 

Proposed Change (Corrected as discussed in the Ad Hoc meeting):

Replace: "An estimate of the offset (in 10 ns units) from the point in time at which the start of the preamble of the PPDU arrived at the receive antenna connector to the point in time at which this primitive is issued to the MAC."

With: "An estimate of the offset (in 10 ns units) from the point in time at which the start of the preamble of the PPDU arrived at the receive antenna to the point in time that the time stamp is generated."

Note this definition should also be corrected at the following additional locations: 3114.37, 3152.6, 3168.42, 3217.6, 3235.42, 3392.27, 3523.10, and 3578.20 (9 locations in all).

 

Note that CID – 3098 proposes to “clean-up” the note where the generation of the timestamp is described in 6.3.55 and that the offset referred to in 6.3.55 is the RX_START_OF_Frame_OFFSET.

To unsubscribe from the STDS-802-11 list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11&A=1

To unsubscribe from the STDS-802-11 list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11&A=1