The non-geostationary design in low-earth orbit is mentioned with not much detail in this article and would generate far less power for a particular receiving l

The non-geostationary design in low-earth orbit is mentioned with not much detail in this article and would generate far less power for a particular receiving location since it only periodically receives power -- so there are various objections to that approach as well. In the case of the geostationary energy being sent through the atmosphere to the receiver -- they mentioned 2000MW of power being delivered -- assuming you use the area of a house (as viewed from top down) as an area to compare against, by some rough calculations roughly 10KW would be impinging upon a house surface that is roughly 1500sqmeters -- so 10KW of microwave energy directed into a house area -- that is non-trivial, so I assume your comparison between wifi antenna's was in reference to the low earth orbit solution -- the geo stationary solution referenced in the article is very powerful and would be in a totally different class from the approx sub 1W range coming from your wifi antennas. So I would be worried to walk through it.