Distance measurement with radio waves

Introduction:
First of all, we want to excuse us for our bad English. (German pupils :D)
We invented a new, inexpensive device to measure distances up to 1.5km (about 1 mile) with accuracy about ±5 Meter (15 feet). The use of radio waves makes it possible to measure without the target being in sight. This means, you can measure distances through whole buildings.There are many rangefinders available, which are working with sound waves or lasers. A disadvantage of distance measurement with laser rangefinder is that you must center up the beam to the receiver and ensure that there are no obstacles along the laser beam.
Schematics and layouts are 100% own work, no copy and paste, only the transmitter and receiver modules had been bought.We already took part with this project in a German youth science competition called „Jugend-Forscht“ and won the 1st prize.

Step 1 Step 1: Basic idea

Step 1: Basic idea
To put it simply, the main part is an exact stopwatch, which measures time with a resolution in nanoseconds. It is used to stop the time the emitted radio wave is travelling. Because the spreading rate of radio waves is identical with velocity of light, you can calculate the distance between the two devices (measuring points) by a given travel time of the radio waves.The stopwatch contains a crystal with a clock rate of 30 Megahertz and a couple of decade counters (High- Speed CMOS). To display the stopped time, binary outputs of the decade counters must be converted to be easier readable on 7-segment-displays. The process of a single measurement:
1) The measurement is being initiated (started with a button) by the user at the basic station (1st point)
2) Counter starts, at exactly the same time a 434 MHz AM transmitter module emits out a 1st radio wave
3) The radio wave gets into the receiver at the 2nd point, and immediately starts the 2nd transmitter at a frequency of 868 MHz
4) The 868 MHz wave is being received at the basic station and stops the counter
5) The travelling time can be read on the display.

Comments

[George.cn]

Hi Jones, I would like to build your circuit but the components list doesn´t match with the schematic, for instance you have 74HC4543 in parts list , I couldn´t find it in schematic, could you add more details please, thanks.

[chouskikou]

Hi guys, I am very impressed with your project, my son is trying to do something similar, do you mind me asking if you were able to purchase the transmitter receiver off the shelf, they need to buy something to achieve the distance measurement, they are in a project to measure the location of a book in a library so they need three of the counters and one receiver/transmitter board to emben within the book. any help appreciated

Chris

[udhayavanan]

hi jones. it is very nice. can you let me more informtion of your project. pls. kindly tell your mail id.

[luogang]

hello, very nice projict. i'd like to coorperate with you to make this project more perfect and even make it a good tool in engineering. please sent me an email [e-mail], looking forward your reply.

[invent4you]

Hello,

Good job!
I would like to build your circuit to use with a project I am working on. Can you explain the use of the 4040 ic's in the schematic?


Thanks,

[JonesiPhone]

Hey there,
tahnks a lot man !
The 4040 are used to count down the delay of the transmitter moduls. As you calibrate the circuit you will get a specific delay of these moduls. With these three counters you can "delete" this delay. It works like this:
While the set count (delay of the moduls) isnt reached is the second Flip-Flop not set, so the count Enable of the 4510 is HIGH.
As the 4040 reached their value the Enable Pin will be LOW, and the actual time of flight will be counted.

That is all. Simple but it works perfect :D
Hope you understood everything an knew now what the 4040 have to do in this circuit

[invent4you]

Hello,

It makes perfect sense. I have been evaluating transmitters and receivers and the issue I am working through, thought wise, is how to account for the differing delay interaction between some of the hardware. Thanks to you, I now know how. Thanks again for your help

Dean

[jwzumwalt]

I have been waiting to see someone tackle a radio range project. Some aircraft use "height above ground" measurement by using phase shift instead of time. A quick short RF burst is fired at the ground and a measurement is made of the reflected receiving pulse's phase shift. This method requires less precise electronics.

It probably would be quite simple for a low power/speed micro-controller to work. I belive there are some common chips that automaticaly calculate phase shift; perhaps out putting a voltage. Wouldn't it be nice if a low cost digital phase chip could be found.

Anyway, you did a nice job, thanks for sharing.

[JonesiPhone]

Of course we know this other method, but they use the reflection. doing so you only can measure the distance to the reflection object. With our project we can measure through most objects without any optical connection, like it needed by laser distance measurement. So this is our advantage.
But your also right, you could solve this problem by using a microcontroller. In our oppinion a microcontroller is much to much for this simple but intelligent project.

[e2f1]

Great piece of work. Can you let me have the supplier and part number for each radio module you used?
Also, did you go through any process of trial and error selecting radio modules until you found modules that would provide deterministic or fixed propagation delays at front and backend stages?

[JonesiPhone]

of course we went through this process. First we used radio moduls out of a garage opener, second moduls had been radio controlls used in RC Cars, and the best one, we use now, are two simple walky talkys build in 1980, which use very old transistors and no bit pattern. These bits and their accuracy is so unusable if you want to count some nanoseconds.
So just look up for some old walky talkys on ebay or so and have fun rebuilding our circuit

[zioelp62]

Cool idea, but IMHO you could achieve a much better reliability with a simple inversion of the principle: instead of sending bursts of waves and measuring the travel time of them, you could send a continuous stream of waves and measure the travel time of suitable "holes", i.e. blank periods. Doing so, the receivers are not allowed to catch RF noise, because they work on a strong, stable carrier. In practice, the sender and the receiver shall broadcast their own signal in short bursts of, let's say, 100 mS on / 1000 mS off, in order to obey the EMC rules about maximum spectrum occupation of 6 minutes per hour. According to your idea, as soon as the sender stops the 433 MHz signal, the receiver recognizes the "silence" and immediately replies by stopping its own carrier @868 MHz. The traveling time is now the interval between the first and the second silences. Furthermore, shifting the measure on the blanks will open some interesting possibilities, first of all the modulation of some ID signals. What do you think about that?

[teddywa]

the schematic in step 5, it just lay out component, could you mail me the wiring schematic and the receiving measurement point 2 also,.
thank you, [e-mail]

[carmaster]

Instead of using all those 74xx chips, why not use an Arduino?
Because it involves programming?

[mobino]

I think because they need to be able to respond to a 30Mhz ripple.
Arduino couldn't - could it?

[Jones Electronic]

maybe they can, but we've got no experience with these "high-tech" chips.
they would be very helpful to backup the previous measurements so we can realize diagramms or such things. so maybe in the future we'll include one

[mobino]

The arduino clock is ~20Mhz. It could be used in place of the drivers to collect data and log it to a PC via serial. It could be used to perform a large number of measurements without manual work. With 3 different measurements you could do on-board trilateration and output 3D coordinates.

Getting an accurate count of millions or billions of ripples is not something any AVR can do by itself. You'd need a high quality TDC for reliable nanosecond accuracy of time of flight measurement.

This project is very good for long distance measurements but indoor positioning (as an example) would benefit more form pre-recorded data for comparison and multiple layers of averaging. I only mention this cause a lot of people would love to have something like this for rover/robot remote control. Unfortunately rf travels a bit too fast. for open field navigation timing ultrasound is a better option.

For me this is interesting for land surveying. I'd love to get this down to 30cm. If you could design a 1ns counter. I'd build it - with arduino logging for 3 bounces. :)

[Captain_Nemo]

GPS is this inverted.

[kurtzthegreat]

it seems to me that if you guys used walkie talkies conected to a microcontroler instead of building your own custom radios you would be able to have greater accuracy (maybe) and a lower cost

[Jones Electronic]

we didnt build our own radio, described on step 5, we bought some and i think the costs, about $75, are very low. another advantage of building it like we did you dont have to programm some microcontroller, in our point of view this way seems to be easier. and the accuracy depends only on our clock signal with 30 MHz, descibed in step 3.
but thanks for your request!

[AvinashPendyala]

@Jones Electronics: dude can u just mail me a schematic with better quality and visibility to this e-mail address "[e-mail]"

[Jones Electronic]

here's the best schematic

http://www.instructables.com/files/orig/F3V/KH3R/GLFUDO8C/F3VKH3RGLFUDO8C.jpg

[AvinashPendyala]

thanks a lot bro.....I owe you one

[static]

Congratulations on your first prize win. As refinement you might consider using sub-audible tones on the transmitters so the receivers would ignore signals that don't have the tone. The chances of other transmitters using the same tones as you do on two different frequencies would have to be very remote. The potential of interference will remain, but I believe their would be reasonable assurance that it was the control unit that triggered the remote unit, and it was your remote unit that the control unit received. Look up continuous tone-coded squelch system on the web, if you don't know what I mean by sub-audible tone.

Anyone here in the USA who desires to build this using the inexpensive 434 MHZ modules be aware though the chance of interference is slight.434 MHZ is in or near a band that the US military still uses for RADAR. 434 certainly falls into an allocation Amateur radio operators have secondary user status use in.

[timdgoodpaster]

It would be good that you sell the PCB. I use Express PCB Express to make my boards
http://expresspcb.com/ They are very fast three days and I have boards in hand. They also the program to design PCB which is very easy to use. This program is free down load Tim

[mobino]

Can you explain your example calculation in more detail?

How do you get 560ns from the displayed 17?

Also, how is a nanosecond timer achieved with 30Mhz?

A 10ns timer could be achieved with 100Mhz (10ns cycle time)

a 30Mhz cycle is aprox 33ns

never mind I answered my own question :)
17 * 33ns = 561ns

I like it

what does the circuit for the far side look like?

[Jones Electronic]

your self-answered question was right 17(displayed) * 33ns (from the 30MHz) = 561. so with the accuracy of this crystal we calculated with 560ns. this is acceptable at a resolution of ±5 meter, we dont care of one meter (1ns) more or less.
we described this as a timer which counts in nanoseconds, not in one nanosecond, sorry for this mistake!
we'll post some new photos from our new circuit and the second measure point soon, please feel free to contact us if you've got more questions

[Jones Electronic]

the first picture shows our second version of the circuit
the second one shows the back side of it
the 3rd one is our second measure point, which receives the signal and sents it on another frequency back
the last one is the backside of our first verion circuit board

[pkiesskalt]

Hi,
A great project.
Your English is very good, but may I suggest one little alteration.
Just call the "die Rueckseite" only "the back", as the "backside" would translate in German roughly "der Hintern".
Greetings from Australia,
Paul K (native German)

[ARJOON]

too complex for me. still cool

[smd75jr]

Could you upload/link to a higher resolution schematic please?

[Jones Electronic]

The schematic is displayed in a low resolution here. But here is the original size available.

[t.rohner]

Click on the little i in the upper left corner, you can open the full resolution image. This way, it's perfectly readable.

To the author: Very nice TTL-graveyard ;-) in a time, everyone uses PIC's or Atmels. (I know, they are to slow for this.) We used to measure frequencies, in thin film vacuum plating machines in a similar fashion. Or more precisely, frequency changes. We had a quarz crystal, which was exposed to the vapour source and changed it's frequency according to the film thickness. We used a 100MHz clock and a faster first stage counter. This way, we didn't count the actual frequency of the crystal, but it's wavelenght. So the change was read instantly. It used a i8086 to connect to a DEC PDP-11 Q-bus.
It had a 7 layer pcb, 25 years ago... i had to test / troubleshoot them with a 100MHz oszilloscope...

[vatosupreme]

Nice project.

Could you possibly add the schematics as a pdf so we could download them? When I enlarged the photo, it was blurry so I am having a difficult time with it.

thanks

[Jones Electronic]

Please guys, all you viewed this project please rate it!

We'll keep you up to date with our testings and developing!

[uberdeity]

Hey, guys- excellent article!

I used to work on underwater positioning systems, and they're basically the same except with sound rather than radio waves. Your system would be called "Long BaseLine" or "LBL" positioning and- using just sound in water- can provide a 3D positional accuracy of about 4mm.

Could I suggest adding a microcontroller on the receiver element? This would let you give each receiver an address and make your trilateration a bit easier as you could interrogate multiple measuring points from a single base station without having to use a wide range of frequencies.
Also, as others have suggested, if you encoded your signal (for example using RPSK) you could get a much more reliable signal.

Also, have a look at using _multiple_ 868MHz receiver elements spaced less than 1 wavelength apart- you would use your current timing setup to get the distance of your measuring point, but you could also use the phase difference between the received signals being received at the base station to calculate the bearing between your base station and the measuring points.
This would allow you to achieve trilateration with only one base station and one measuring point (this is called USBL, or Ultra Short BaseLine positioning).

Hope these ideas help!
UberDeity

[Countrywings]

Really interesting article!
Keep up the good work!

Starliner

[teddlesruss]

Older aircraft DME (Distance Measuring Equipment) used a precision ramp voltage generator, sent a pulse and started the ramp generator, the base station received the pulse, waited a fixed few microseconds and then sent a reply pulse, and the DME stopped the ramp generator and the distance was proportional to the voltage. It was, if I recall, pretty accurate fro its time. (Which was early 70's) The amusing thing for me as a young trainee was the name of the high impedance precision ramp generator - it was referred to in the circuits as the "phantastron" circuit. I found out that this is a legit name, BTW.

http://www.answers.com/topic/phantastron

[Infinitevortex]

That's fantastic! Good job.

[geonav1]

Your project is amazing! Wonderful work- please keep going and continue to share your experience.

Wunderbar!