Basically, given no GPS is allowed, can we use walkie talkie to locate the position of a car / a person?
I am assuming the walkie talkie is using radio, not GPS-based.
What equipments will be required to capture the signals in and out to approximate the location (within 2 miles range)?
Thanks.
Basically what Jeff said.
With a single receiver, the best you can do is to estimate someone's direction. But if you move to different locations, or have two receivers, then you can triangulate.
The question is: what do you really want to do?
This might be possible with something that no longer resembles a walkie talkie much. With a highly directional antenna, and the ability to measure time of flight of the signal, you could probably work out someone's heading and distance.
This is easier, and you only need to make the antenna directional. Ham radio operators have direction-finding contests called "foxhunts", and have designed extremely effective antennas (and techniques) especially for that purpose.
If your walkie talkies are in the roughly 460MHz band, then many of those antennas will be suitable. Whatever your frequency band, I'm sure you'll be able to find a suitable directional antenna design somewhere online. For example, here's a WiFi directional antenna:
Finding the person is simply a matter of working out which direction to walk in, and walking that way. However, one thing that messes this up are reflections. In hilly country, you might find yourself taken on a wild goose chase.
Trilateration can be performed with a single receiver if it can move and the transmitter is stationary. One would have to compare the received signal strength at 3 or more (multilateration) locations. This comparison must be done to signals transmitted at the same (or known) power. If the walkie-talkie pair uses amplitude modulation, this means the transmitter needs to send the same tone and amplitude -- no sound input might work. Frequency modulation schemes can include a carrier or squelch tone, transmitted at a constant amplitude. This method can be simplified if the transmitted power is known.
Radiated radii is about \$r=\sqrt{P_\text{avg}}\cdot 10^{dB\mu/20+1}\$. By overestimating the transmitted power by a bit one can graph the trilateration circles. This example uses a 22mW estimation for a 15mW transmitter.
This graphical method can be extended to multilateration, using a least squares technique on an overdetermined system to estimate the transmitter location.
Using trilateration with received power readings won't give accurate results, especially when line-of-sight is obstructed, but will give a general area in which to search. How accurate is it? Maybe one of these papers can tell you, as that question is complicated: Efficient Solution and Performance Analysis of 3D Position Estimation by Trilateration, A Simple and Efficient Estimator for Hyperbolic Location, A Passive Localization Algorithm and Its Accuracy Analysis, Recursive Position Estimation in Sensor Networks.
