Dear all,
I am new in k-wave, basically I am performing simulations with an initial dose deposition from Geant4 which is then converted into initial pressure. This 3D dose deposition is generated with a dx = dy = dz = 200 um.
Is it mandatory to use the same grid resolution (dx,dy,dz) in k-wave? What is the best way to choose my grid resolution?
Best.
Hi,
Basically, yes. To the best of my knowlegde, k-Wave only accepts the same resolution along the three dimensions, although the team was working on nonuniform fft a few years ago (I don't know where they are on this matter right now).
I do not work on initial pressure problems so I can't help you on the choice of the grid resolution. Maybe if you do a fft of your initial distribution, you should be able to determine the smallest wavelength (thus, the highest frequencies) you need to properly define your initial distribution...
Hope it helps,
Hi Anthony,
""""Basically, yes. To the best of my knowlegde, k-Wave only accepts the same resolution along the three dimensions, although the team was working on nonuniform fft a few years ago (I don't know where they are on this matter right now).""""
Where did you get this information ? I have never read or heard this restriction. I have used k-Wave for 5 months and I have often used different resolutions along the 3 dimensions. I have compared my results with theory many times and no problem... I'm really astonished about what you're saying...
Best regards,
Alex
Hi All,
There is no specific restriction in the code to fix dx = dy = dz. However, keep in mind that if you use a different grid spacing, you will have a different maximum supported frequency along each axial dimension. This may give you strange results, particularly for broadband sources. In general, I'd advise keeping them the same if it's feasible. However, I could imagine some situations where it's advantageous for them to be different. For example, if you have a step interface or thin layered structure aligned with the grid in one of the dimensions, it may be beneficial to decrease the grid step in that direction to improve the resolution of the layer or interface.
Brad.
Thank you all, and I am sorry for the late reply. Nevertheless, I am using a uniform grid spacing to avoid problems.
You must log in to post.