This is nonsense. If the power level is high enough to do damage, detecting it would be trivially easy.
Just a wire dipole, a diode, and a simple meter will do the job. At most it would require a set of dipoles to cover each band of interest.
Most radio hobbyists could make a set of dipoles from an a old coat-hanger in a few minutes.
Further if you knew what you were trying to do, you could set up your feedback transmitter in a room, record all the parameters for that room, then move to the next room, Etc.
The problem that Perlman was solving was a lot harder (a moving target) but hitting a nominally stationary target?
I guess semiconductors are covering pretty high frequencies these days though(http://www.semiconductor-today.com/news_items/2019/may/knu_2...).
Remember, bog-standard corruption in Ukraine is a crazy conspiracy theory which should be banned from social media.
But mystical microwave illnesses? Very credible.
You have permission to re-engage your bullshit detector after all the votes are in.