[time-nuts] question for expert time guys

Chris Albertson albertson.chris at gmail.com
Mon Feb 4 14:11:19 EST 2013


I think for best timing rather then use a "ping" Where the device gets
a signal, waits then sends a reply it would be better to re-transmit
the signal in real time.   So the base station send some waveform like
carrier that is modulated by a saw tooth wave.   The remote device
simply re-transmits the sawtooth wave but on a different carrier
frequency.   Then the base station looks at the phase difference in
the transmitted vs. received saw tooth.    You can get even more fancy
and make the sawtooth something more complex such as random digital
data and use an auto correlation filter to find the delay.  but phase
works if you select the signal such that there are no range
ambiguities (in other words such that phase is never over 360 degrees)
   You can even start with a long period (none ambiguous) saw tooth
then when you know the approximate range more to a shorter period.
You can think of better waveforms, like FM modulation or a block of
digital data.

This works because (1) Each cycle of the saw tooth is the same as your
"ping" and you can average many thousands of them and (2) There is
zero turn around at the remote end.  It is just a "bent pipe" and you
don't need a computer.   The "bent pipe" is low cost.  Only one site
needs to be expensive that others just analog reflectors

In fact people have use corner cube reflectors for range measurement.
Basically to hit the reflector with a radar.  If you want low cost I
can't think of anything cheaper than a few sheets of aluminum.
My sugestion about was actually an "active" reflector or
"transponder".  It can make the radar cheaper because less power is
required. ANd you don't need to turn off the transmitter to receive.
If you are limited to milliwatts, you will need the active
transponder.

In short, think of this as a kind a radar and it becomes a lot less
complex and it can even be 100% analog with a UV meter use to display
the range (or phase of received signal)




On Mon, Feb 4, 2013 at 9:35 AM, James Peroulas <james at peroulas.com> wrote:
> About this subject, are you concerned with multipath? The signal from two
> of the basestations might arrive over a line-of-sight path whereas for the
> third basestation the signal might bounce around before arriving...
>
> JP
>
>
> -------------------------
>
> To time experts/EE's.
>
> I would like to triangulate a position of a device which moves using 3
> fixed positions devices of known location.
> The idea is to have these operate on 915mhz or 434mhz or 2.4ghz or
> appropriate frequency.
> These two type of devices (fixed and mobile) are all under my control and
> thus customized as needed.
>
> The mobile device (not a phone, custom device) would be the least expensive
> item.  I'd like a range of 150 feet or better and accuracy of 3 feet or
> better.
> When manufactured these items they can be calibrated in order to adjust for
> any variation in IC's, discrete components etc...
> We can assume for now the temperature is constant 70 degree temperature.
> Cost is the key design factor.
>
> The general flow is:
>
>    1. base station 1 indicates we are determining position of device A.
>    2. Each base station 1, 2, 3 take turns pinging the device to determine
> distance.
>    3. A ping consists of (something like, e.g. frequencies as examples)
>         -send 915mhz signal from base station to device
>         -device response ASAP on different frequency
>         -station waits and counts 'time' for return
>         -this is repeated N? times to get best avg/accuracy.
>         -The mobile device does not move very fast
>    4. Since delays of the process on each unit is calibrated the device and
> base station would subtrack that time out from the results.
>    5. obviously with 3 distances we can determine the 2D position of the
> mobile device
>
> I know the time accuracy is the key to count time =  feet, 1ns.  This
> overall project is not new concept.   How to make it "inexpensive" is key.
> how inexpensive, very ;-)   no OCXO or expensive components like that.
>
> That's my goal, and I'm looking for help on the design/thought process of
> getting there.
> I am open to a consulting arrangement for a fee, please email if you like.
>  I've worked with 'regular' EE's (I'm a software guy) but this time
> accuracy is too much for them.
> Esp. finding a way to do it inexpensively.
>
> Thanks for any thoughts.
>
>
>
> --
> Integrity is a binary state - either you have it or you don’t. - John Doerr
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



-- 

Chris Albertson
Redondo Beach, California


More information about the time-nuts mailing list