Re: [STDS-802-11-TGAZ] FW: RE: [STDS-802-11-TGBD] TGbd and positioning
Hi Ganesh,
_____
From: *** 802.11 TGaz - NGP - Next Generation Positioning ***
[mailto:STDS-802-11-TGAZ@xxxxxxxxxxxxxxxxx] On Behalf Of Venkatesan, Ganesh
Sent: Friday, November 6, 2020 12:45 PM
To: STDS-802-11-TGAZ@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-11-TGAZ] FW: RE: [STDS-802-11-TGBD] TGbd and
positioning
Hello Dick:
Your messages are interesting (as always).
[RR] Well, that might be a bit of a stretch, but thanks for the kind
sentiment :^)))
In .11az (as in REVmc Fine Timing Measurement Protocol) our positioning was
always relative to 'known' geo-spatial co-ordinates of the AP(s) against
which the positioning protocol is executed. If the positioning protocol is
executed against 3 or more APs (which have been administratively configured
with their LCI co-ordinates), we could triangulate and estimate the LCI
co-ordinates of the STA that executed the protocol against each of the
AP(s).
That is how the protocol works; and we have verified that it works as
expected by testing with conformant (IEEE802.11-2016) implementations.
[RR] Let me be brief, and if I have misunderstood what you said and what
.11az is doing, pounce on me!
All of what you say certainly can be done, and I have no doubt you have
simulations and experimental evidence to prove the point (I have done
thousands of such simulations and real-world experiments in the last 4
decades so I am sure you can as well). However it is exactly what I was
suggesting below should be avoided if possible at all cost. Data
preprocessing is going on and, especially in this case, a ton of valuable
information is being lost/destroyed. Given that the objective is to produce
a "best estimate" of the "location of a STA", I can assure you the "optimal"
thing to do is make the time measurements (TMs), attach the measurement
error (co-)variance and all the necessary metadata to make those
measurements unambiguously interpretable (which includes the estimates of
the locations of the APs and their estimate error covariances among other
things) and pass them along to a "kinematic and attitude state estimator
(KASE as I call it :^)))" where optimal statistical signal processing will
produce outputs that are the "provably best estimates" of the things I want
to know given ALL the information it has at its disposal (and the
assumptions made in setting up the problem but that's another story for
another day). What more could you want?? :^)))
Sorry to be the bearer of such news, however at least I am not lying to you
:^))) Like it or not, this is the "truth" :^)))
Cheers,
RR
Cheers --
ganesh
"It is amazing what you can accomplish if you don't care who gets the
credit." - Harry Truman
From: *** 802.11 TGaz - NGP - Next Generation Positioning ***
________________________________________________________________________
To unsubscribe from the STDS-802-11-TGAZ list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-11-TGAZ&A=1