Homebrew detector?

[xwarp]

but if i remember my experience with the patriot system in el paso myself, the patriot anttenna is phased array. with this kind of set up, the polarization of the received signal is kind of a mute point, but with a radar detector, your already losing signal power because of the detectors feedhorn.

hey walt, shoot me an e-mail.

[The Walt]

but if i remember my experience with the patriot system in el paso myself, the patriot anttenna is phased array. with this kind of set up, the polarization of the received signal is kind of a mute point, but with a radar detector, your already losing signal power because of the detectors feedhorn.

hey walt, shoot me an e-mail.

Yes, it's phased array radar. Built for longer and more precise targeting and correlating the beam patterns.

[stalkervision]

but if i remember my experience with the patriot system in el paso myself, the patriot anttenna is phased array. with this kind of set up, the polarization of the received signal is kind of a mute point, but with a radar detector, your already losing signal power because of the detectors feedhorn.

hey walt, shoot me an e-mail.

actually you are quite right...smart of you to see this! 8) There is also the problem that most traffic radar is "circularly polarized" to pick up radom microwave polarization's coming from off of your vehicles since your vehicle was never designed to be a radar reflector. This probably cuts down on a detector's received signal even more! One wonders why the radar detector companies don't take this into account and design an antenna that is circularly polarised also into their products. My theory about this is that the shape of the antenna would have to be a cone as traffic radar is and this would look none to pretty on a radar detector.... :wink:

[kai445]

It would take quite a bit of work to make your idea work well.

You could get things like the horn antenna and stuff, but the real tricks will be in the signal processing. It is like your radio... you hear lots of static, but you might hear a peep here and there. Your DSP (or analog) would need to analyse this signal and see if it is real. The Escort/Bel products seem to do this better than the V1. However, it takes quite a bit of experience to determine real/fake signals. Plus, KA band will be the killer, as it is a large sweep.

Sorry for bumping this old thread but I'm still (at least mildly) interested in this. So I have a few questions that I think would be crucial to explore.

Do you think a chip as simple as a dsPIC30F would be powerful enough to analyze the signal? I'm not really familiar with the PIC family, but they seem to be popular with hobbyists doing somewhat simplistic applications... but the DSP models are relatively new and seem more unexplored.

...or does anyone know of a more advanced chip for prototyping in an application for this. I'm always open to ideas.

Here is a brochure: http://ww1.microchip.com/downloads/e...doc/70095k.pdf

These things are only like 6 to 20 bucks (in quantities of 1, at mouser).


PS - My original vision was a multi-component detector, not nessicarily something stuck in a windshield. I'd like to be able to interlink components ("brain" unit, laser diode assembly, radar antenna, remote display, whatever) with standard CAT5 cables, if at all possible.

PPS - I have no intention of reverse-engineering, modifying, decompiling, or otherwise touching an existing radar detector, I don't care if you want to, so don't even ask. I have no intention of breaking or infringing on any patents either, so don't even think about it. I wanted this to be a "homebrew" project, not a "rip off an existing company" project.

[jimbonzzz]

The chip itself might be cheap. But, have you considered the cost of the development tools? I think a detector with such digital processing would be in very difficult reach of the homebrew hobbyist...

[kai445]

The chip itself might be cheap. But, have you considered the cost of the development tools? I think a detector with such digital processing would be in very difficult reach of the homebrew hobbyist...

Ah, good question.
The "general purpose development board" from Microchip (the makers of the dsPIC) is $299. Other companies have programmers between $119 and $239. There are probably more inexpensive solutions also.

The DSP analysis/processing software (dsPICworks) and DSP libraries are actually free (downloadable from here. The IDE and Assembly "compiler" is free also. The C compiler is expensive (like $895). So homebrew would have to be done in ASM unless there are inexpensive or free C compilers available.

I'm keeping my eyes open for other DSP solutions. If anyone else knows of one, dont hesitate to mention it :-).