All radio receivers have local oscillators built into their receiving circuitry that actually transmits energy that can be detected by specialized receivers.
Not only are there radar detector detectors, but then there are radar detector detector detectors, and radar detector detector detector detectors. A hilarious arms race.
> All radio receivers have local oscillators built into their receiving circuitry that actually transmits energy that can be detected by specialized receivers.
All efficient radio receivers....
It is possible to build radio receivers without such. I built a crystal set as a child that could power an air piece without a battery using only the AM signal.
Not so much with modern ADC based receivers, it is possible that the only oscillator is at a fixed frequency and the only “tuning” happens in a computer.
It's basically just a more sensitive radar detector. Instead of picking up the main frequencies radar guns use, it picks up the harmonics emitted by the oscillating crystal inside the radar detectors. All radar detectors leak on harmonic frequencies of the frequency they pick up on, it's just a matter of how much. Those cheap Walmart radar detectors leak so badly they detect each other. Pricier ones are more sophisticated in how they shield.
Could one build an SDR version of a radar detector that doesn't have a detectable oscillator? (Or, if it does, it's oscillating at a much higher frequency?)
I'm no RF guy, so I'm clueless on this stuff. But my understanding is that SDRs aren't radio receivers in the traditional sense. That's the whole "SD" part of it, the device scoops up everything and FFTs it or something.
But they are suffering from the chip shortage.
I was interested in using it as an X-band and K-band capable SDR receiver, which is a supported use case according to their media.
But we'll see when it ships.
It would be way cheaper to design the radar detector to have better RF isolation than to go towards an SDR solution.
A modern radar gun operates in the Ka-band, Google says between 33.4 and 36 GHz. Assuming 12 bit samples, your proposed solution needs to process a minimum 3.9 gigabits per second, or ~500 megabytes per second. To do this work in software would require a workstation.
Not to mention I'm not even sure if you can operate most ADCs above the 1st Nyquist zone like would be needed here.
> To do this work in software would require a workstation.
These days hardware capable of that throughput can be had for less than $500 and draws less than 50W. 3.9 gigabits really isn't all that much by modern standards.
You could tune an antenna to trip a detector detector (side note: which would be some great civil disobedience!), but I don’t think you can get to the detector detector detector without an interaction with law enforcement. Layer eight problem.