Polarity preserving decon in ``N log N'' time

GETTING THE BEST OF BOTH WORLDS

The good news is that it is possible to have the best of both worlds, both to deal time-symmetrically with the Ricker wavelet and non-time-symmetrically with the bubble. Furthermore, we do it rapidly in fast Fourier $N\log_2 N$ time. The Ricker wavelet is quite short, about 15ms, while the bubble delay is quite long, about 150ms, so the separation of the two is not delicate.

The less happy news is that the required theory is quite deep and not well known. It was first invented by famous mathematicians about 1940, and did not reach textbook status until my 1974 book FGDP, with code included in my newer book PVI 1992. Luckily, I have been teaching this material for many years now and believe I can extract the essence without great pain for you.

Polarity preserving decon in ``N log N'' time