uncleJim21 on Nostr: Maybe but you could put in some neat tricks to brick the device if that's tried. I'll ...

Maybe but you could put in some neat tricks to brick the device if that's tried. I'll have to think on that. You could also make it a MEMs or a specialized IC that's part of an inline (high side current) circuit that would brick the device if removed or altered. The reason this attack could work is because STMs are so common they're like the Smiths of electronics.

This is my working design. I would also go ape shit and try to minimize power consumption of the device even if it means making you wait 10s instead of 3s for it to do a cryptographic operation. Makes it easier to detect fuckery.

Channel A detects instantaneous power if they're just total hogs and go for it (RF is incredibly peaky)

Channel B mitigates risk of them charging up a capacitor or some other energy storage and trying to surreptitiously charge up for the Big Transmit.