Hi Joel,

Joel Kolstad wrote:
Christopher wrote:

Hi guys,

ok, I've got an idea but it's based on determining the distance of a
transmitter from a receiver, I originally thought about a synchronised
clock in both units, the transmitter sends the time it has out, by the
time the receiver unit gets this time a period has passed (probably a
few millionth's of a second) and a diff time is detemined, combined with
the speed those waves travel, will reveal the distance.


Yes, but as you've deduced, the problem is keeping the clocks synchronize.

On the other hand, there's no reason you can't have one unit send out a
pulse and begin timing. The second unit receives it, immediate sends a
'reply' pulse, and you stop your clock when the first unit receives that.
Now the accuracy is determined by the accuracy of just one clock, the amount
of uncertainty in the 'turnaround' time of the second unit... and the
variation in the speed of light as a function of athmospheric conditions.
(The last one being more of a concern if you're trying to listen to
satellites -- GPS receivers run into this problem.)

my problem with this is that radio waves are electromagnetic radiation
which travels at the speed of light, therefore can travel 1 meter in 3.3
nanoseconds, if my range is around 10 meters (I was a bit conservative
before, lets say 30meter width/depth and 10 meters high, that'd be big
enough I think for any situation I'd use this system within.

Is there a clock out there that can measure accuracy on this scale, it
seems impossible. That's of course, if radio waves travel through our
atmosphere at almost the speed of light, I'm going to look now as to how
fast they travel through our atmosphere.


You can also use multiple transmitters with high-quality clocks and one
receiver -- this is what GPS goes. The transmitters send out a message
saying, 'I sent this at exactly this time... really I did!' You record on
your local clock when you receive those messages and hence can perform
triangulation to determine your position. Better yet, if you have yet
another transmitter, you can solve for your position as well as the
_relative error in your timebase_ and thereby synchronize one (or more)
receivers to the high-quality clock in the transmitter. In GPS receivers,
this additional satellite is what gets you a 2D fix vs. a 3D fix.

well if you have one transmitter attached to say the top of your head,
you'd need multiple receivers to be able to triangulate that
transmitters position wouldnt you, you'd record the time it was sent and
diff it against the time you have in your local clock, that difference
would give you a rough distance to the transmitter, from each receiver
you'd have a distance/transmitter value, which would denote the RADIUS
of a circle from which the transmitter could be, of course, the location
of the transmitter is where all three circles overlap. More receiver
units could be used to increase the accuracy of those measurements.

the problem again is clock sync. I did originally think that you could
have one master receiver unit, whose clock all others are determined by,
this clock could send out it's time with a "reset" code that all the
other clocks would reset their time to, this would happen fairly
accurately would it? I think it would. Then you're clocks are fairly
sync'd. You'd probably not want to run them for long without resync'ing
again though


signal strength perhaps?
literally I am talking about a transmitter within a cuboid shaped
enclosure around 10m maximum and being able to pinpoint that transmitter
within that enclosure accurately, to around 1cm, perhaps 2cm.


Ah, OK, if it's enclosed field strength would be a reasonably reliable
indicator. There was an article in Circuit Cellar Ink just a year or so
back where a guy did this (and cited references -- the idea has been around
for awhile), but it's sort of the oppposite of what you suggest: You stick
large (transmitting) coils at the edges of your box and then detect the
signal strength on a small receiver. 1cm in 10m is 1 part in 1000, which I
would gander is right about in the middle of 'trivial' and 'pushing
impossible.'

hmmm, thats interesting idea. although I can't find much about it,
could you remember the article name?



There are also '3D' computer mice (mouses?) out there that use a couple of
ultrasonic receivers on the 'mouse pad' and perform triangulation after
listening for the mouse's transmission. Same idea as with RF, but easier
because sound waves are so much slower than electromagnetic waves -- times
are measured in microseconds or milliseconds rather than picosecond and
nanoseconds!

---Joel Kolstad