> I don't know what I could see.

I would imagine at a bare minimum the continual but stochastic presence of RF signals in certain frequency ranges, and at those ranges harmonics, might indicate computational activity as opposed to just a plain or stuck draw/sink.

Computation is usually fairly noisy, encryption moreso.

Non-linear junction detectors operate on that principle for detecting hidden audio bugs or cameras by passive listening, or actively by directed radio (looking for the characteristic reflections).

There are also some projects like DeepTempest that seek to actually reconstruct remote activity within a device based on RF emissions. A lot of insights though will depend on how well you know the hardware.

Anyway, this was just a thought. Too many mfg's today are doing stuff they shouldn't be doing with their equipment to the consumer's detriment. You really almost need to have an EE degree just to figure out what it is they are doing, or even know specifically what's happening in most cases.