My counter-counter-intelligence idea is that since you know where the radio broadcast station is, you can probably implement some kind of faraday cage with an aperture oriented towards the broadcast station. That way any kind of radiation emitted from the receiver also gets sent in that direction. That way detecting a signal emanating from the receiver requires being in the cone outside of that aperture (and I doubt those signals are so strong that they can be seen from space).

I'm sure your counter-counter-counter measures might start from the premise that, with you also knowing the location of the broadcast station, you might be able to guess at the most likely receiving orientations if you knew the most likely places for a receiver to be located. Eg a shortwave reciever equipped with a faraday cage + aperture listening to codes from Russia in Manhattan would emit a cone in the direction of Stamford. But I also suppose that I might know my receiver emits this kind of signal, and use some other device to emit junk signals that look similar, or scatter around a bunch of receivers while only using one.

Or I might try to only set up receivers in places where the cone would be inconvenient for you to intercept, or obvious that you were trying to intercept. Eg over the ocean.