[aprssig] Re: Indoor tracking

[Chris Kantarjiev]

(I get the digest, so I'm collapsing a lot of responses into one message.)

I would like to start out by saying that I worked with a fairly
large deployment of the Olivetti Active Badge in the early 90s.
(It was IR based, transmit only, sending a beacon with the unit
ID and one-bit of data: 0 if a timed beacon, 1 if you pressed the
switch; the second generation allowed some 2-way.)

There is no doubt that you can do a lot of cool stuff with indoor
location and willing participants. But if you think that trying to manage
600 digipeaters across the US is tough, wait until you try to deploy
something to all 600,000 ARRL users

> But... you don't need the RFID tags for this.  The "trackees" are already 
> carrying a transmitter that beacons its name periodically.  ...
> Why not just detect the 144.39, read the id, and shoot that out.

This seems like a really good idea to me, requiring a little extra
smarts in the digipeater that is mounted at the HAM clubhouse or
wherever, and utilizing the existing APRS infrastructure. It might
overwhelm the RF channel, unless the traffic went straight to TCP.
Such a smart digipeater might even send ?aprs? every once in a while
to encourage a beacon from everyone.

It's not entirely clear to me, though, why it's necessary to track
folks while they're in the meeting, since the last posit from their
vehicle is probably at the parking lot. What am I missing here?

> We need the HUGE code space so that we
> can MAP our callsigns to the built-in codes.
> with a single reproducible algorithm.  THe way
> I see it, for ARRL calls, the code space has to
> be A,N,K,W for the first letter (4)
> then  4 characters of 26 possibilities then one
> character of 10 possible numerals and one digit of
> 3 possiblities to tell where to put the numeric.

I don't really believe it works this way. I think RFID tag information
is allocated in a similar fashion to Ethernet MAC IDs: you purchase
an allotment of a unique address space and have to map it. In the
simplest case, it is then up to a computer behind the reader to
do the mapping of the tag ID to the identity of the item. But,
obviously, there are many many schemes in place. (RFID is *far*
from standardized, despite what the hype would lead you to believe.)

> So, someone's earlier suggestion that we
> simply have each tag respond with a positionless APRS string was an easy
> implementation.

Yes. This seems great to me - it could even be a self-powered tag
this way, avoiding battery issues. It's then up to the reader
(or the attached computer) to provide the location information.

> Create a fake GPS satellite to be mounted above each doorway that
> streams out the GPS coordinates of the room you are in. 

I'm afraid I don't understand this at all. The bitstream that comes
from the satellite constellation is completely different than the
GPRMC sentences that come out of the serial port...

> At a conference I attended last year in Dallas, everyone was outfitted with
> an intelligent "badge" which was more of a PDA with 2-way 900MHz radio
> inside. ... This badge did its work
> using strictly RSSI - I'm not sure how it dealt with multihop (if it even
> did deal with it).

This sounds incredibly neat. If we're already talking about installing
new receive hardware, I think it would be great to do this kind
of solution on 144.39 MHz.

I bet it's incredibly difficult to do well, though.

>  What's the goal here?

I'll echo this sentiment. And ask, in addition - other than the intended
audience, what does an RFID-based solution have to do with Amateur Radio?

73 de chris KG6VYD