sensing vibrations

The ray paths of seismic rays depends on the geological setting of the subsurface.
So one possibility is to use geological information, the other is to filter the seismic information with respect to surface waves and treat them as linear.
What is the approach?
BTW: nice to be here

Hi McN,
nice to have you here.
You are basically saying, that the different subsurfaces have different capabilities to transfer seismic rays?
How significant would these different subsurfaces affect transmission speed / capabilities?

The approach for now is, to sense vibrations over a longer distance. These vibrations are produced by machines, turbines, etc.

Depending on seismic velocity of the surface and average gradient of the velocityfield (say dv/dz) the traveltimes of the rays are significantly higher. It might be that the surface velocity is 1500 m/s and only 50 meter below the velocity is 3000 m/s. If we now use the surfacevelocity but the ray propagate in the faster layer below we have a huge error in the position of the source (or the receiver). It is possible to filter the waveform with respect to surfacewaves. They are also not linear but the approximation is much better.

Sorry for the late reply.
I assume that not the time is the most important, how long it will take from sender to receiver, but the signal strength is more relevant (With this principle is Google's wifi positioning working).
How much do you think will the signal strength related to distance be affected by different surfaces?
Do you have any tests, references where somebody has tried this or something similar?

The signal strength depends highly on the length of the propagation path and the length of the propagation path depends on the material of the subsurface. But I am sure that estimated information about the subsurface are sufficient for a good approximation of the decreasing amplitude of the signal
in a certain distance.
Full waveform inversions in the field of seismics or seismology have the same problem. The advantage here is that you have days or weeks time for calculations. So you can create a model of the subsurface and so the exact position of source/receiver. The best way in our case is to reject the subsurface by filtering the surface waves.