There is no such thing as immunity to RF interference.
When one transmits a signal, that signal does not exist only at that frequency. There will be attenuated signal peaks at harmonics of that frequency (multiples). There may also be many spurious peaks at other frequencies, which are a function of the circuit design and installation. Additionally, there isn't really a fequency (other than 0) at which one could say the power output is truly zero. It is extremely close to zero for most of the frequency range.
When one receives a signal, the receiver is not sensitive to only that one intended frequency. A receiver has a "passband" which is a (usually) narrow range of frequency over which it is essentially equally sensitive. A receiver tries to reject frequencies that are not in this band. As one scans frequency outside the passband (
https://en.wikipedia.org/wiki/Passband), the sensitivity of a receiver drops off fairly rapidly. But it does not drop off to zero. It drops off to the level of sensitivity which is in the "stop band" (
https://en.wikipedia.org/wiki/Stopband) which may be only 50dB (dB -> decibel, or tenth of a Bel... this is a logarithmic measurement scale for signal strength and for power levels
https://en.wikipedia.org/wiki/Decibel) down from the pass band sensitivity, or even less, for cheap hobbiest and consumer level electronic circuitry. Really good receivers may be in the 100dB or higher rejection range. But that is still not perfect.
When one applies a signal that is outside the designed passband, a receiver will reject that signal to the extent that it can. If a receiver is receiving a signal at a level 'A', and is rejecting a signal at level 'B', B can give a stronger reception result than A if B is at a sufficiently higher level than A - enough to overcome the stop band rejection level.
Signal strength is a function of the distance between the transmitter antenna and the receiver and its antenna. The received power level drops off with the square of the distance. Distance is your friend!
Orientation matters as well. If a simple transmit antenna is oriented at 90 degrees to a simple receive antenna, this will substantially reduce the received signal, due to the difference in polarization.
But in the near field, where transmitter and receiver are within a few wavelengths of each other, things get more complex. When they are very close to each other, they can couple fairly strongly. A transformer would be an example of such coupling being used as a feature.
One should be careful of how one defines a transmitter and a receiver.
A transmitter is nothing more than something putting an RF (radio frequency) signal over a wire. That wire can be a circuit trace on a board.
A receiver is nothing more than a wire connected to something capable of dealing with signals. Again, that can be a circuit trace on a board.
So, there are many places in a circuit which can transmit RF, likely at many different frequencies and likely with many different signal characteristics. There are many places in a circuit which can receive RF, again with many different reception characteristics.
For receivers, places which contain amplifiers in a circuit are possibly the biggest risk.
Portions of a receiver circuit can transmit; portions of a transmitter circuit can receive.
You can try putting the electronics in a Faraday cage, but a transmitter or a receiver has to have an antenna outside the cage. Even for non- transmitters and receivers, you may need access for power, and for altimeters, there needs to be an air hole or a barometric sensor won't work (not via trivial designs, that is, but there are ways such as a metallic diaphram to transfer pressure while maintaining the cage, over some moderate pressure range). One could put the altimeter with its battery in a Mu metal box (
https://en.wikipedia.org/wiki/Mu-metal) to shield against RF and magnetic fields (another way to transfer signals in the near field). One would have to do the engineering for transfer of pressure and control of on-off function.
I hope this helps reduce confusion rather than add to it. For those who know what I'm talking about, please pardon me for trying to avoid most any technical term, to hopefully not lose readers.
Gerald (KE4ET)