Using RF for relative direction finding for short ranges (sub 10m)

Question

I'm considering building a short-distance direction finder to determine the relative position of two objects.

I want to be able to determine direction through other objects, which eliminates IR or ultrasonic detectors. This leaves RF.

Antenna design I've seen seems focused on long distances and on quality of signal. However, my goals are short range (less than 10m) and direction. I don't care what gets transmitted, just where it comes from.

I expect that one of the objects would have an omnidirectional beacon and the other object a directional antenna. An angle slot of 15 degrees would be acceptable. I can deal with multiple antennae.

There's various antenna configurations that look like they'd be sufficient for directionality. They are: Yagi-Uno, dish and HB9CV-Beam. However, I don't see how to shrink them, or which ones are suitable for shrinking to short distances and small receiver sizes.

I expect I'll have to build custom mechanical and electronics hardware, although off-the-shelf parts would make my life a lot easier.

Here are some specific questions:

• Which antenna configurations are suitable for shrinking?
• I think I need a very small wavelength, so would be looking in the 2Ghz range. Is that right?
• What power output should I be considering? Is this at all possible? Are there examples of this?

I have knowledge of electronics but am obviously new to antenna design. I appreciate any answers you might have.

Answer 1

The most effective antennas rely on the length of the "parts" being a defined sub-fraction of the wavelength of the signal that is to be received. This is usually a quarter wavelength (dipoles) for both halves or juts a quarter wavelength for a one-ended dipole (aka a quarter wave dipole).

Yagi antennas use other fractions to define the length of the reflector, driven element and directors.

Parabolic dishes become more directional and have increased gain as frequency rises but these will need to be big to get the directionality I think you need.

In short, I think you may be fighting some of the laws of physics.

Some good news - you won't be needing more than a milliwatt given the distances you are talking about.

Answer 2

Have a look at Texas Instruments mmWave technology. It is different than what you described, a transmitter and a receiver. It is a radar system so it shows objects and not a specific object. But if nothing else, it will show you the benefit of going even higher in frequency than you considered.

http://www.ti.com/sensors/mmwave/iwr/overview.html Videos at the bottom of the page.