Link to home
Start Free TrialLog in
Avatar of jtiernan2008
jtiernan2008

asked on

measuring time delay between RFID wireless reader and tag

I just picked up the following of ebay for a project....
http://cgi.ebay.ie/ws/eBayISAPI.dll?ViewItem&item=290198500910&ru=http%3A%2F%2Fshop.ebay.ie%3A80%2F%3F_from%3DR40%26_trksid%3Dm38.l1313%26_nkw%3D290198500910%2B%26_fvi%3D1 
I am trying to develop a multilateration real time location system based in differential time of arrival.

Can someone please advise what is the best way to measure the time taken from reader to senser also what is the required apparatus?

Thanks in advance.
Avatar of aburr
aburr
Flag of United States of America image

I think you have a difficult problem. You will need to measure a very small time difference VERY accurately.
I suspect that you will need the accuracy of an atomic clock. Some are very small, all are expensive.
The stability of the frequency will have to be great if a phase difference is used. The stability of the start time would have to be great.
ASKER CERTIFIED SOLUTION
Avatar of d-glitch
d-glitch
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of jtiernan2008
jtiernan2008

ASKER

=> This hardware is probably not capable of measuring time delays in the nanosecond regime.
But I thought the core purpose of the device is just to broadcast and detect the signals. The detection of the signal would be up to extra hardware or software reading the ports? I have written a simple java programs that reads the ports and puts a time stamp using "long startTime = System.nanoTime();"

=>The carrier frequency is 315 MHz, but the communication link operates at 9600 baud.
This is like a 20 year old modem.  
I actually got the 433 Mhz version if that makes any difference.

=>You need approx 10,000x better resolution to get 1 ft spatial resolution.
By resolution, I thought dealt with the frequency that the tag beacons and is read. So if it beacons and is read faster then the reproduced co-ordinates are updated more frequently. Would nanoseconds not be neglible since this is faster than is noticable by the human eye... I may not be understanding this correctly.

=>The transmitter operates over a range of 10-20 meters, and itsounds  like it sends and ID string of four
8-bit characters, probably every 10 seconds or so.  It runs on a watch battery.  

It sends a signal every ~1sec.

=>It takes on the order of a milliisecond to send this sort of string.

So this has in impact on the accuracy/resolution?

=> You would do better with a transmitter that send a 1 us pulse, maybe every second.
You mean the speed of broadcast is 1 microsecond every second... but it does broadcast every second. In addition, why does the transmit speed need to be 1us if it is only supposed to broadcast every second?

=>The receivers all need to be have fast response detectors, linked to a high-speed counter (in the 200 MHz
to 1 GHz range).  This is not necessarily expensive, but it is very specialized equipment.

What do you mean by specialised equipment and where would you get this equipment?

Thanks for your help so far :)
Try looking at this problem in 1 dimension:

You have two receivers 10 meters apart.  
You have a 300 MHz transmitter that sends out a 1 us long pulse once every millisecond.

If the transmitter is halfway between the receivers, the differential delay will be zero.

If the transmitter is 6m from RX1 and 4m from RX2, the differential delay will be 2/(3x10^8) = 6.67 ns.

If need to see spatial resolution in feet, you MUST have time resolution in ns.

At 300 MHz, the period (one full carrier cycle) is 3.3 ns.  It is difficult/impossible to reliably detect a carrier in less than 1 period.

GPS systems use much higher fequencies (1575 and 1227 MHz) and only guarantee resolutions on the
order of 10 meters.  

        http://www.gmat.unsw.edu.au/snap/gps/glossary_i-q.htm

If you want to do position monitoring with a PC, you might want to consider ultrasonic transducers rather
than RF ones.  The speed of sound is approx 1000 ft/s, so millisecond resolution is just fine.

If you want to use RF techniques, you probably should look at Differential GPS.

     http://www.trimble.com/gps/dgps.shtml

Thanks a million for your extremally helpful advise so far...

Just two things..

=>You have a 300 MHz transmitter that sends out a 1 us long pulse once every millisecond.
 1 us long pulse - Is this 1 microsecond long pulse?

Also you are seem to be a very good expert in the area!! I think I need further research into the area of clocking, fast response detection and resolution and accuracy...Are there in any text books you would recommend from experience that may cover these concepts?

Finally... you said above, This is not necessarily expensive, but it is very specialized equipment
What do you mean by "specialized" equipment and can you give me some examples of this equipment?

thanks a million for your help so far.
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Its a project so the resolution will not really matter I suppose and I can't purchase new equipment....
Thanks d-glitch for all your help... I will get working on this!!!