[drjesusphd]

Just to clarify, this is not an intentional design feature. Any reciever is necessarily a transmitter due to the physics itself.

The UK has been using this for a while to crack down on TV liscences.

[joezydeco]

I'm skeptical about the License Van thing...

"BBC admits that TV detector vans only work because Britons believe they do"

http://gadgets.boingboing.net/2008/10/31/bbc-admits-that-tv-...

[drjesusphd]

Interesting, I didn't know that. So they're more like polygraphs.

[jrockway]

With a good isolation amplifier between your antenna and local oscillator, you don't necessarily transmit much on the frequency you're listening to. (If you did, you'd interfere with other listeners.) Where you transmit also depends very much on the design of the radio you're using; it might be obvious to detect that a radio is being operated, but not detect what frequency it is tuned to.

And to be very pedantic, every conductive object in the universe plays some part in any antenna system.

[nitrogen]

Don't most receivers use a standardized intermediate frequency, though, resulting in leakage that always leads back to the original frequency?

[jrockway]

Yes, most radios use standardized IFs.

But some high-end SDRs use ridiculously-high sampling rates (like 200MHz), in which case many applications don't even need to mix a local oscillator with the incoming RF. It's pretty amazing.

[nitrogen]

I never learned much about RF, so this question may be naive. Is it possible to sample higher frequency signals using a 200MHz sample rate, by using a bandpass filter and then deliberately undersampling without an anti-alias filter? If so, how high could you go, and what other limitations exist?

[coin]

There are also radar detector detectors, operating off the same principe - detecting the local oscillator of the receiver.