# Triangulating Location

Discussion in 'General Electronics Discussion' started by gamenut, Sep 22, 2011.

12
0
Sep 22, 2011
Hey all, I am working on a new project where I need to get a position, basically in millimeters, using 4 transmitters and a receiver. Right now the problem that I am having is that to triangulate position there are two methods that I know, first off you can use signal strength (my preferred one) to determine is, however I am not entirely sure what hardware is commonly used today to get a measurement of signal strength. The other is based on time, basically however long it takes for the radio wave to reach the receiver. I am working in a small room, about 50 meters, and want to get an accurate measurement. Considering radio waves move at the speed of light, I can't get a good timer that can respond within microseconds. So basically here is the question, does anyone know another method of doing this? And if not, is there a way to implement the timed based method? Also, where can I find a signal meter? Thanks

2. ### poor mystic

1,074
36
Apr 8, 2011
Hi
Signal strength is not a good way to determine position. Noise from outside sources intereferes unpredictably with the signal and ruins the fine measurements that woul be required. Also, radio signal strength does not fall off linearly with distance between transmitter and receiver, especially close to the transmitter, and that's another trouble.
The best method that I know of is actually the 'time' method.

One of the variations of the time method involves phase-locking onto the signal, and comparing the derived, phase-locked signal with an accurate clock within the receiver unit. Of course this is not easy to accomplish. Many variations of time-based position determination exist.
.
The most important factor in deciding on the method is the degree of accuracy required in the result.

I wonder why you are using 4 transmitters? What's it all for? What frequencies are these transmitters? How stable are their clocks? Whar errors are acceptable?

I have so many questions I cannot begin to write them down, for fear of spending my whole day on this post, and I hope you soon explain the problem in more detail.

12
0
Sep 22, 2011
I am using 4 transmitters because as far as I know that is the best for 3 dimensional triangulation. All I want to do is access a objects position in a 3d space relative to the 4 transmitters. As for frequencies or clocks, that is something I haven't worked on. If you can't tell, I am actually a programmer, but I need some hardware for my next project.

The main issue is this, I am trying to get a 3 dimensional position using the 4 transmitters like satellites and the 1 receiver like a gps receiver. However they use radio waves, which is perfectly fine if your object is in space, but it gets to being a much more precise system within a small range. If you can point me in the right direction as far as clocks and transmitters go, I should be able to work it out. Sorry for the lack of detail

4. ### poor mystic

1,074
36
Apr 8, 2011

However good a programmer you are, you're going to have to get a bit more in touch with physics. Time is distance.
Accordingly, the exact frequency and phase of radio waves received at a receiver, having been transmitted from several stable transmitters at known locations, can be used to locate the receiver in space to a precision that can be discussed at a later date.
Only 3 transmitters are required for 3-space work.

Last edited: Sep 24, 2011

12
0
Sep 22, 2011
Hey, thanks for the quick response. Sadly, my experience with code hasn't really helped me with this. The main problem is that radio waves transmit at about 300,000,000 meters per second, and I want to test in milimeters, or 300,000,000,000 mm per second. This is a bit to fast for a normal clock as far as I know. Any ideas?

6. ### duke37

5,364
772
Jan 9, 2011
You can use sound waves to generate longer time intervals. Sound ranging was used in the first word war to determine the position of the enemy's guns. Pity the poor sod who had to go out with the microphone on the end of a wire. I believe that the same thing is done these days with microphones shot into position and using radio reporting. Early TV remote controls used ultrasonics.

Doing it with signal strength would be very difficult with directional tranducers and reflections, maybe possible in the open or anechoic chamber.

7. ### poor mystic

1,074
36
Apr 8, 2011
the most important question is still What errors are acceptable?
Your
mention of millimetres suggests that you want to know within a millimetre where your receiver is. Is that really what you want?

443
3
Jun 10, 2011
I understand your desires for such a system, as I have wanted such a thing for decades for a variety of purposes. One clue to the difficulties should be that there are no consumer systems available, even if the cost were allowed to get up into the k\$ region (at least to my knowledge). Time of flight is too demanding on timing, which is why you don't see the method being used by distance measuring equipment) and acoustical techniques suffer from reflection problems in real-world environments. GPS might be usable, but it suffers from signal strength problems indoors and the need for significant averaging times to get cm accuracies.

9. ### poor mystic

1,074
36
Apr 8, 2011
I like what you said, daddles, about wanting such a device. We all want one.
Gamenut, I should have made it clear that I thought the project impractical, from the start. The project can't happen on the scale you have in mind.
However if you could be content in a much larger space, say the sky, there is technology to fix your position with useful precision..

10. ### poor mystic

1,074
36
Apr 8, 2011
Do you think this could be done with light?

Well, I think it could be done with light but what good would it do you? Unless your game is to be played at the bottom of a mineshaft. To use light would mean either having 3 pointable detectors on the receiver that could track the apparent positions of 3 fixed lasers; or working in total darkness, so that you don't need the pointability in order to detect the light..

Last edited: Sep 24, 2011

443
3
Jun 10, 2011
Hi, Mark -- how are you?

Well, with enough work, it could be done. You could put a device on that would use laser distance measuring technology to measure distances to known points in the space. That technology is used in laser distance meters and total stations; it's mature and capable of measuring out to 50-100 m or more. You'd have to have some way of tracking the known points when the measuring platform is moving. This could be done with e.g. IR emitters that blink a unique code; this would let the point be identified by the platform. Then that point would have an omnidirectional reflector that would let the distance be measured to it.

So I think in principle it could be done. But this would mean there can't be any optical obstructions to the known points. Or, there have to be enough known points to allow you to triangulate your position from anywhere, in spite of the obstructions.

So: doable, but it's not going to be cheap and it will be a fair bit of development work.

12. ### poor mystic

1,074
36
Apr 8, 2011
Hi Don
Yes I'm quite satisfactorily happy having been much occupied in various pursuits to get me off the internet. The net can be an unhealthy timesucker for me, with too much time on my hands. So I've been out-and-about, meeting people and doing things.

There is one other technology I can think of that could help.
A radio signal could be bounced back and forward between 2 transceivers functioning as repeaters. The rate at which one of the transceivers re-transmitted the signal would be a function of distance which could be calibrated. Thus, the distance between the transceivers could be measured.

12
0
Sep 22, 2011
What would be the issues with using sound, like duke said? Sound waves travel much slower, but vary depending on temperature and the material they travel through. As for rebounding through transceivers, that also seems doable.

443
3
Jun 10, 2011
You should do some homework first -- e.g., google "acoustic location" and learn about things that have already been done over the last century or so.

12
0
Sep 22, 2011
Wouldn't it be possible using radio waves to measure how many times the tranceiver received the transmission per second, basically do a rebounding effect like poor mystic suggested. If an object is say 1 meter away, then it receives 300,000,000 transmissions a second. If it is 10 meters away, then it receives 30,000,000 transmissions a second. And just link this to a counter to determine position? What do you all think?

16. ### duke37

5,364
772
Jan 9, 2011
The TV shows repeats of sports events using Hawkeye to show the path of a ball. I have no idea of the technology involved

17. ### poor mystic

1,074
36
Apr 8, 2011
Last edited: Sep 26, 2011

12
0
Sep 22, 2011
Haha, well that's good to know, now it's time for the actual implementation. Research time! And thanks for the link poor mystic, I will give it a look soon.

19. ### davennModerator

13,983
2,015
Sep 5, 2009
yup but incredibly expensive !! hi speed video cameras dont come cheap, they can run into 10's of 1000's each 100's of 1000's for the real serious ones

that was an interesting article

cheers
Dave

12
0
Sep 22, 2011
I'm sorry, but I'm confused on the use of a high speed camera in this situation. Finding the location using a camera seems like a common situation in, say, augmented reality programming using images on paper. This is a bit different. Would it be possible to consistently send out a signal, then send back a signal to the transmitter, and repeat the process.

For example, you have 2 transceivers, each sending back and forth signals for about 1 second. The transceivers are about 10 meters apart, and no objects are between them. Considering the speed of radio waves (300,000,000 meters per second), wouldn't the transceivers be able to transmit around 30,000,000 times in that second. So couldn't you have a counter for these transmissions and determine that it was transmitted that many times because of its location.

You could also measure in milliseconds for millimeters.

Now, what would be the big problem in this?