Simple continuous tracking radar system with cantenna - how to get ranging info?

[VinnyCee]

Simple continuous tracking radar system with cantenna

We are in need of a simple radar target tracking signal processing method. It's to be very short range, less than 5 meters. It needs to track a single target that can be moving or stationary at any time. We are using 2.4 GHz ISM band.

Here is an MIT OCW site about the hardware that we have to work with on the RF end...

MIT Online Courseware - Small Radar System

We have the hardware all assembled and working right now.

We have also purchased an Arduino MEGA 2560 for signal processing.

We also have a video output device with which we made a couple of games the radar system will control. Kind of like a Playstation Move device, only using radar and probably much less functional.

So what type of signal processing should we do on Arduino in order to provide tracking or ranging information? The target (a large metal-covered apron which the user wears) can be moving or stationary at any time. The range of detection only needs to be a few meters, nothing more.

I've looked into http://www.scribd.com/doc/17533697/Chapter-2-MTI-Radar and it seems its all geared towards outdoor use, crazy amounts of targets, etc.

Can anyone direct me to some resources that have done this? If not, what would you recommend in terms of signal processing?

 

[jsgruszynski]

First, a CW transmitter won't track Az, El or Range. You can get velocity but that's about it. You can integrate velocity to get distance but it has all the calibration problems of an accelerometer with regard to distance (i.e. it drifts and often badly due to acceleration accuracy errors and numerical integration errors).

You generally must have phasing in the antenna (see "monopulse radar") for Az/El resolution/measurement and you need some pulse modulation for ranging. Resolving a single object isn't as easy as it seems because every reflecting surface (walls, people, etc.) all look exactly the same as your target, either in return time (ranging) or in AzEl space. Because you can't really get an arbitrarily small "pencil beam" (like a laser pointer), you can't be sure about what is really returning the signal without more complexity.

 

[skeptic2]

...you need some pulse modulation for ranging. .

This is not true. If the transmitter were frequency modulated with either a triangle or sawtooth wave, the reflected wave will lag the transmitted wave and will have a slightly different frequency. Mixing the transmit frequency with the reflected signal will give a difference frequency proportional to the distance.