k-Wave

1. qiangbo
   

   Member
   Joined: Jul '13
   Posts: 15

   Is it possible to mix nonlinear and linear materials in the same phantom? Thanks.

   Posted 12 years ago #

2. Anthony
   

   Member
   Joined: Apr '13
   Posts: 96

   Hi,

   I am only a k-wave user (it is therefore not an "official response") but I would say that it is possible by using an heterogeneous "medium.BonA" parameter with 0 in the linear material. However, the calculation will probably be done by k-wave with the nonlinear set of equations, even in the linear part of the phantom, what should slow down the calculation compared to a linear simulation.

   Once again, my answer is only based on my personal experience of k-wave, the development team will probably give you a more precise answer :-)

   Anthony

   Posted 12 years ago #

3. Bradley Treeby
   

   Developer
   Joined: Mar '10
   Posts: 1,643

   Hi Bo,

   If you define medium.BonA, the nonlinear equations will be used (as Anthony mentions above). However, even if you set medium.BonA to zero in some regions of the domain, the model will still include convective nonlinearity as explained here. It's implemented this way as this makes the most sense physically.

   Having said that, if you are using the MATLAB code and want to switch off the nonlinearity in regions where BonA is 0, open up the simulation functions and find where rhox, rhoy and rhoz are calculated at the next time step (around line 900 for kspaceFirstOrder3D in V1.0). Then for each line in the nonlinear case, replace

   rhox = bsxfun(@times, pml_x, ( bsxfun(@times, pml_x, rhox) - dt.*rho0 .* duxdx ) ./ (1 + 2*dt.*duxdx));

   with something like

   rhox = bsxfun(@times, pml_x, ( bsxfun(@times, pml_x, rhox) - dt.*rho0 .* duxdx ) ./ (1 + (medium.BonA ~= 0) .* 2*dt.*duxdx));

   This will force the convective term to be zero whenever BonA is zero.

   Hope that helps,

   Brad.

   Posted 12 years ago #

RSS feed for this topic

Reply

You must log in to post.