The primary problem with this concept is that you have to know very precisely the composition of the ground where you install this barrier. Another problem is that environmental changes - soil moisture, temperature, are going to affect the material properties somewhat (but maybe not enough to matter).
Essentially, extremely low frequency waves that trash buildings don't perceive the ground as atomic, the waves act over their wavelength, which is very long, and so if you put things into the ground, it changes the material properties. Carefully drilled holes apparently can change the properties in dramatic ways. The word "cloak" is sexy, but the more interesting bit mentioned at the end of the paper was the prospect of building a bandstop damper with the low corner at 0 Hz.
It doesn't do you much good if your earthquake prevention device reflects the energy somewhere else dependent on the epicenter, and it also doesn't do you much good if it doesn't block enough frequencies to stop it from trashing your buildings. A bandstop filter would operate over a broad enough band to attenuate all the frequencies, and it wouldn't reflect energy to other buildings (which could have obvious liability concerns.) Imagine a plaintiff's attorney showing a standing wave pattern of destruction emanating from a field of holes drilled by the defendant's firm.
The other satisfying nature of this tech is that it's proactive. Instead of building structures that will probably collapse if a magnitude 8 happens anyway, you go out there and build armor that will stop the earthquake entirely. Also, a field of holes and concrete and various pertubations, all buried, is a lot less ugly than the structural changes needed to reinforce a building against a major earthquake.
It would be expensive to do the detailed surveys and compute the solution, but it would create more high education jobs, and it's probably worth doing.