Well, the nuts and bolts of the system got immediately classified, so there’s really only rumor and guesswork as to how it actually works.
That having been said, from my admittedly very limited knowledge of microwave transmission and detection, I have a couple of ideas how it might work.
A) The system actually detects “void” areas. Instead of sending a search-radar beam out and getting a reflected signal, the cell array makes note of “empty” areas, where there should be signal, but there isn’t.
If Tom Clancy can be believed, some of our modern submarines suffer the same problem- they’re actually quieter than the surrounding random noise of the ocean, and so can be located (roughly) by looking for small, localized quiet zones.
B) The cell arrays are so sensitive, they can detect even the tiny fraction of reflected energy.
Or C) The arrays are just that, arrays. “Stealth” technology doesn’t necessarily “absorb” all the microwave energy, most of it is simply reflected away from the point source. With a widely-seperated array, other cells can detect the energy reflected from different directions
Point-source search radars have the emitter and detector at the same- or nearly so- location, so the stealth has to reflect the energy off to the side, or above and below. But with an array, there’s additional “detectors” off in the directions the energy is being reflected to.
Personally, I think it’s a factor of the last two. The cell receptors are typically dealing with fractional-milliwatt signals, so they have to be pretty sensitive to keep a good voice signal. (Which is surely a more complex signal than simply a mere reflected fixed pulse-rate signal from a search-radar.) And second, the arrays are wide-set, heavily interconnected and designed to share and transfer signals as the user moves from cell to cell.
It’s actually a brilliant idea. How long before the FAA starts leasing air-time from the Baby Bells to augment their 60’s era traffic-control radars? 