Distance Finding with RF Ping

Thread Starter

tehmarkster

Joined Feb 11, 2011
2
The team I'm working with is trying to design a position finding system that would work in an area with a 25m radius with a relatively high resolution. What I've determined is that to get the most accurate results you want to do this with a ping (i.e. not using signal strength).

We've looked at using an ultrasonic pulse, but it has distance limitations and in particular the system is meant to be in someone's yard so those might bug animals.

The idea I'm on right now is that whatever object is tracking, we place an RFID tag on them. We can then triangulate the position by sending an RF signal, and timing the delay from sending the signal to getting a response from the RFID tag. To get accurate results we'd need to be able to measure the delay to the nearest ns or better. If the timer is accurate to the nearest ns, that would tell us the distance within .15m of the RFID tag which would be acceptable. Are there any timers out there that has this level of accuracy?

Also I found a patent for a system using a similar method here: http://www.freepatentsonline.com/5828443.pdf.

My only concern is that the counter we'd need to use would run at 1GHz, but we want to keep the cost down low (that's if we can even find such a counter).
 

thatoneguy

Joined Feb 19, 2009
6,359
Radar isn't simple or cheap (essentially what you are building)

The speed of the circuitry would need to be much higher than 1Ghz to process the 1nS pulse time (1nS per foot, roughly). 1Ghz is the pulse repetition rate, the receiving and processing would need to be faster than that.

Have you looked at the higher end sonar modules (only 15m range though)?

Laser Rangefinders would seem to fit your needs, but you may have "sticker shock" when looking at the price tags.

RFID tags don't have a 25m range, need to be close enough to the transmitter to get an induced current in order to transmit, which also causes an unknown delay in your current idea.
 

Thread Starter

tehmarkster

Joined Feb 11, 2011
2
So based on what you've said I wouldn't even consider radar for this solution. Ideally we wanted to find the distance to a passive device, but it is not necessary that it be passive so RFID isn't necessary either. We considered using ultrasonic range finding, but we were concerned about the range and its effects on animals since it'd be used outside. Also I was concerned about the directionality of ultrasound. We would not be measuring the time between reflections, but instead we'd do a call and response in order to increase distance.

So essentially my question is this. If we had a system that had two ultrasonic transceivers. One would send out a ping and the other would respond. Would the orientation of the two devices affect the performance. I.e. do they need to be facing each other. Also would this system be feasible in an open air environment say in somebody's backyard.

Also, you mentioned laser range finding, but I don't think that'll work since we do not change orientation like a moving robot. I didn't read the full article and I will do more research into laser range finding, but at first glance I don't think it'll work.
 

jpanhalt

Joined Jan 18, 2008
11,087
Stanley Tools used to sell a laser ranger finder for about $50. That's what I paid, new-in-box on ebay. It uses time of flight like the radar analogy. It is quite accurate (<1/4"). Brand name was FatMax, which is applied to a variety of other tools too. It is far superior to the ultrasonic "tape measures."

John
 

someonesdad

Joined Jul 7, 2009
1,583
Stanley Tools used to sell a laser ranger finder for about $50. That's what I paid, new-in-box on ebay. It uses time of flight like the radar analogy. It is quite accurate (<1/4"). Brand name was FatMax, which is applied to a variety of other tools too. It is far superior to the ultrasonic "tape measures."

John
John, it seems the manufacturers like to keep quiet about what technology they use (or are barred because of NDAs), but I doubt that device used time of flight. Here was an explanation I wrote up after a bit of reading the web a few years ago when someone gave me a laser distance meter:

The laser beam is capable of being modulated in amplitude. Internally, the laser beam is split into two signals, one that is kept inside the instrument and the other which is sent to the target and reflected back to the instrument. The frequency of modulation is changed and the phase between the internal beam and the reflected beam are compared. When it's known that a phase shift of no more than 2π radians has occurred between the two beams, then the difference is due to the time it took for the reflected beam to travel. This phase difference of the modulating frequency is then converted to a distance.

Since the speed of light in vacuum and air is about 3e8 m/s, light will travel 0.3 m in 1 ns (a nanosecond is one-billionth of a second, or 1e-9 s). To travel 1 mm (the unit I have is accurate to about a mm for modest distances) takes 300 times less time; you can see that exquisite timing would be needed if this type of instrument worked on the time-of-flight principle. Instead, this phase shift method is used, which shares some characteristics of the famous Michelson interferometer used in the Michelson-Morley experiment.
 

jpanhalt

Joined Jan 18, 2008
11,087
Phase shift is clearly related to TOF. But, given the wavelength of light, there needs to be some sort of timer/counter to distinguish, say 10 ft from 20 ft. I called technical service and talked to an engineer. He confirmed in writing (e-mail) that it was TOF. You are free to do with that information as you wish.

I believe the design was Swiss based. It's cheap and doable. I am aware of the speed of light and the constraints that places on timing.

John
 

terra

Joined May 12, 2011
13
The RF travels at the speed of Light, so if you need to measure the distance less than 50m and the accuracy of 0.15m then you need a timer(similar like stop watch) at the accurracy of ns, then you need at lease 50Mhz or more than that low PPM crystal or you can create the clock using FPGA. Why don't you think about GPS with RTK signal
 

Wendy

Joined Mar 24, 2008
23,415
I like the phase discriminator for fine measurements, it eliminates the high precision counter. Instead, you get an analog measurement that can be converted to a digital measurement. The counter would be for the coarse measurements, the multiples of the TOF, and the phase discriminator would tweak it to a decent accuracy. I hadn't thought of it in that light before.

If it is radio, and used for range finding, it is RADAR. I'm not sure what the OP thinks RADAR is, but that is the original definition (Radio Detection and Ranging).
 

hgmjr

Joined Jan 28, 2005
9,027
I suspect that the vernier method is being used to accomplish the TOF measurement in most of the instruments commercially available.

Link: Vernier applet

Before there were digital readout calipers, we used the mechanical vernier type similar to the one simulated here. Such calipers were able to achieve very fine resolutions.

If you can come up with a circuit that mimics a vernier type measurement then you will be able to make the measurement you seek with the resolution you are trying to obtain.

hgmjr
 

Wendy

Joined Mar 24, 2008
23,415
It can be, but it doesn't have to be. There is a difference. The base definition does not mention distance as a requirement.
 
Top