the need for very accur4ate clocks is that it is in orbit at high velocities. timed pulses could be sent locally with a few known location beacons to do the same thing.

 

And how do the known locations know what time the signal was sent?

 

Same way the satellites do. A local GPS is theoretically possible but I don't think it's practical. Maybe for a permanent installation.

 

...........................
GPS is very expensive technology, it needs very accurate clocks, hence why no one has developed a local version.

Can't be that expensive. Many cell phones have built-in GPS

 

He was referring to the time-synchronized transmitters, which would be costly even if they weren't in space.

 

AFAIK The satellites contain atomic clocks accurate to a billionth of a second or so. A set of base stations would have to do the same. The satellite sends out a time and the position it was at at that time. The receiver reads 4 or more satellites and from that it work out it's position and an error factor which is the error of the inaccurate receiver clock.

You are measuring the delay from the signal being sent to it being received. Light travels 0.3m in a billionth of a second so you are foot out for every 1 billionth of a second that your transmitter clock is out by.

 

then why not use three duplex trancievers? measure the phase difference of a audio tone from a duplex repeater, the delay (phase shift) gives you the range. three such ranges (one for height) will give you the location. of course ignoring the position behind you.

 

Trying to get my head round that, I take it you are talking about setting up an audio "standing wave" on an RF signal. How do you know how many whole cycles you are separated by? What is the wavelength of such a standing wave (my first guess would be the same as the speed of sound in air)?

 

no, just a tone modulated on gthe carrier at the other end, is a duplex repeater, sends out what it recieves. measure the delay (phase shift ) between the outgoing audio and the recieved audio. incoming audio delayed by distance (round trip) years ago when I worked on digital weather radar, one mile was 10.4 micro seconds, I dont remember if thats regular or nautical miles. on zero crossing of outgoing wave, start a clock, a 9.3 mhz clock will give you pretty good resolution, stop count on zero crossing of incoming \wave. distance should be on the counter.

 

I believe RFID is capable of real-time 3D positioning...

Not without GPS.

Finding x, y, and z, without the aid of GPS or the Russian equivalent GLONASS, will be difficult at best.

 

one mile was 10.4 micro seconds

Let us forget about the things that slow down radio waves ...

One statute mile is 5280 feet. Speed of light is approximately 1.016 feet per nanosecond. That makes the one way trip from transmitter to receiver approximately 5.36434 micro seconds.

Things to consider:

equipment delays
secondary phase factors (delays caused by transmitting in the atmosphere)
additional secondary phase factors (delays caused by the propagation path)

Otherwise it's a simple calculation.

Inexpensive would not be in the vocabulary.

 

Not without GPS.

Finding x, y, and z, without the aid of GPS or the Russian equivalent GLONASS, will be difficult at best.

It can be done without GPS, but requires a coordinate system setup around the area to be scanned. I don't know how practical that would be in a Fire scenario though, especially at a high rise building, unless it could be deployed in a quick, safe, and adaptable way.

Basically, the RFID targets would be the Fireman wearing the RFID tags and the 3D coordinate system would be reference RFID tags deployed in a cube-like shape forming the area to be scanned. The system would scan the target area for the RFID wearers..

eT

 

can i use an accelerometer to estimate which floor the firemen is at ? i noe there is three axis in the accelerometer, can i use the z axis to identify when the firemen is moving upwards?

 

Yes, but I think precision will be difficult to achieve. You might experiment with a FitBit. This is a commercial device that does what you want - it estimates how many steps you have climbed using "state of the art" equipment and algorithms. OK, it's not really SOTA, but a compromise to be small and affordable. But anyway I think the algorithm would be difficult to duplicate on your own.

 

With one accelerometer sensor there still will be ambiguity (rotational motion) of direction of travel so you also need a gyroscope for inertial tracking.

http://aerostudents.com/files/avionics/InertialNavigationSystems.pdf

 

Pretty sure the FitBit type devices don't have a gyroscope, although they may have a barometer. Not sure. I may be confusing it with my daughter's Garmin that has a barometer. But her FitBit definitely tries to detect elevation change due to stairs or hill climbing.

 

Pretty sure the FitBit type devices don't have a gyroscope, although they may have a barometer. Not sure. I may be confusing it with my daughter's Garmin that has a barometer. But her FitBit definitely tries to detect elevation change due to stairs or hill climbing.

I think it's got an altimeter/barometer chip inside also to detect up/down changes.
Something like this: http://media.digikey.com/pdf/Data Sheets/Freescale Semi/MPX4115A.pdf
https://learn.adafruit.com/downloads/pdf/fitbit-force-teardown.pdf

 

Ah yes, I recall that teardown. My daughter's first FitBit malfunctioned and that teardown is what convinced me to not bother attempting one myself.