http://www.k-wave.org/forum/topic/boundary-conditions-instead-of-pml Support for the k-Wave MATLAB toolbox en-US Wed, 13 May 2026 09:10:00 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/boundary-conditions-instead-of-pml#post-7517 Thu, 21 May 2020 15:08:53 +0000 leakerb 7517@http://www.k-wave.org/forum/ <p>Thanks Brad! </p> http://www.k-wave.org/forum/topic/boundary-conditions-instead-of-pml#post-7509 Thu, 21 May 2020 11:05:46 +0000 Bradley Treeby 7509@http://www.k-wave.org/forum/ <p>HI leakerb,</p> <p>If you want a periodic boundary condition, just set <code>&#39;PMLAlpha&#39;, 0</code>. The gradients in k-Wave are calculated by an FFT which assumes the field is periodic.</p> <p>You could also see <a href="https://arxiv.org/abs/2005.00322">this paper</a> and <a href="https://github.com/ucl-bug/matlab-dtts">this respository</a>.</p> <p>Brad </p> http://www.k-wave.org/forum/topic/boundary-conditions-instead-of-pml#post-7508 Wed, 20 May 2020 17:03:53 +0000 leakerb 7508@http://www.k-wave.org/forum/ <p>Hello,<br /> I am trying to replicate the results of a paper where instead of the PML they used Dirichlet periodic boundary condition at the edges of the computational grid, so that opposite edges are mirrored.<br /> The authors give the boundary condition for one pair of edges as:<br /> sigma(0,y,t) = sigma(Lx,y,t), 0 &lt;= y &lt;= Ly<br /> v(0,y,t) = v(Lx,y,t), 0 &lt;= y &lt;= Ly<br /> Where Lx and Ly are the width and height of the computational grid.<br /> How would this be implemented in k wave? I've read on some other posts that you can define Dirichlet sources at the edges to enforce boundary conditions, but I'm not sure how to do it for these conditions.<br /> It seems to me like this is essentially the effect of setting PMLalpha to 0? But the way the conditions are described in the paper make me doubtful and I'm pretty new to this.<br /> Thanks! </p>