Here is some math regarding the IFFT.
I'm sampling at 2MSPS, so each horizontal bin represents a signal delay of 500 nsec, or 149.9 meters. With delays of ~50 bins, this comes out to 7.49 km difference in some of the X and O signal that I'm seeing.
As such, I'm looking at a highly magnified view of the ionosphere (difference between X and O signal paths) than others who are looking at a single channel compared with an idealized FMCW waveform.
David