https://www.k-wave.org/forum/forum/general Support for the k-Wave MATLAB toolbox en-US Wed, 17 Jun 2026 21:50:15 +0000 http://bbpress.org/?v=1.0.2 q https://www.k-wave.org/forum/search.php https://www.k-wave.org/forum/topic/complete-beginner-question#post-9247 Tue, 17 Feb 2026 12:48:38 +0000 xf20228 9247@https://www.k-wave.org/forum/ <p>Hi, </p> <p>I am working through the Homogeneous Propagation Medium Example but get this error when I run the example .m file:</p> <p>Arrays have incompatible sizes for this operation.</p> <p>Error in kspaceFirstOrder2D (line 920)<br /> ux_sgx = dt .* rho0_sgx_inv .* real(ifft2( bsxfun(@times, ddx_k_shift_pos, kappa .* fft2(p)) )) / 2;<br /> ^<br /> Error in example_ivp_homogeneous_medium (line 67)<br /> sensor_data = kspaceFirstOrder2D(kgrid, medium, source, sensor);<br /> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<br /> So far, I have installed MATLAB R2025a, installed k-wave from add-ons, opened the example_ivp_homogeneous_medium.m file in the MATLAB editor, and run the code.</p> <p>If someone could tell me what I am doing wrong, I would appreciate it. </p> https://www.k-wave.org/forum/topic/thanks#post-9244 Wed, 28 Jan 2026 17:12:35 +0000 toehead 9244@https://www.k-wave.org/forum/ <p>I just wanted to say thank you for such an amazing toolbox. </p> <p>It was just what I needed and is very user-friendly. </p> https://www.k-wave.org/forum/topic/transducer-object-for-elastic-simulations#post-9228 Fri, 15 Aug 2025 15:27:02 +0000 abbasK 9228@https://www.k-wave.org/forum/ <p>Hello,<br /> can the kWaveTransducer be used with the pstdElastic3D function? </p> https://www.k-wave.org/forum/topic/running-thermal-simulations-on-a-gpu#post-9209 Fri, 18 Apr 2025 09:05:21 +0000 jorgeT 9209@https://www.k-wave.org/forum/ <p>Hi newmam,<br /> For me it is working perfectly. Just copy the code from github as you may have the kWaveDiffusion version that has not yet implemented this feature.</p> <p>JTP </p> https://www.k-wave.org/forum/topic/time-between-saving-steps#post-9206 Thu, 10 Apr 2025 08:59:08 +0000 Dominique 9206@https://www.k-wave.org/forum/ <p>This topic remains pertinent today: I'm required to execute an extensive 2D simulation, yet the storage demands for sensor_data are restricting my progress since it records data at every time step.</p> <p>The solution suggested here is effective; however, I'm apprehensive that similar modifications might be needed in the C++ source code to avoid having to choose between computational speed and excessive RAM usage. </p> https://www.k-wave.org/forum/topic/on-the-change-of-medium-properties-over-time-in-k-wave#post-9199 Sun, 16 Mar 2025 09:00:26 +0000 wjh 9199@https://www.k-wave.org/forum/ <p>Hello, I would like to simulate the effect of time-varying medium changes on acoustic wave propagation properties using k-wave. How do I solve this problem and does k-wave support the simulation of time-varying sound velocity and density of the medium in question? For example, for 2D simulation, do I need to change the internal function of kspaceFirstOrder2D, and if so, how do I need to change it? I hope to get your reply, thank you very much. </p> https://www.k-wave.org/forum/topic/wavefront-index-discrepancy-in-3d-sensor-masks#post-9171 Sat, 11 Jan 2025 00:19:35 +0000 egekckkmrc 9171@https://www.k-wave.org/forum/ <p>Hello,</p> <p>I’m working on propagation in a homogeneous 3D medium. I’m using the kspacefirstorder3D function to accomplish this. I have a specific issue that I need your assistance with.</p> <p>Let’s consider the following scenario: I create a grid with Nx=32, Ny=32, and Nz=32. Then, I generate a linear binary sensor mask array at the first layer of Nz (sensor.mask(:,:,1)=1). I position a point-like source 3 layers below my sensor (Nz=4) and propagate the wavefront, collecting the data. Let’s refer to this data as data1.</p> <p>Next, I repeat this process by creating a sensor mask for the first two Nz (sensor.mask(:,:,1:2)=1). I maintain the same source, perform the propagation, and collect the data. Let’s refer to this data as data2.</p> <p>After some reshaping, I obtain the following for data1: [Nx,Ny,kgrid.Nt] and for data 2: [Nx,Ny,kgrid.Nt,Nz]. I anticipate that data1 should be equal to data2(:,:,:1). However, when I compare them, even though the wavefronts have the same shape, the indices differ.</p> <p>Could you please provide some insights into this issue? I would greatly appreciate your help! </p> https://www.k-wave.org/forum/topic/thank-you#post-9166 Sat, 21 Dec 2024 19:16:11 +0000 Alberto Madeira 9166@https://www.k-wave.org/forum/ <p>I didn't find a proper venue for this, but I really wanted to thank the authors of this toolbox. This work is wonderful! I work with jet aeroacoustics and having a FAST wave propagation solver with such reconfigurability, ease of setup, and powerful numerics behind it is incredible! I've designed countless mechanisms and devices throughout the (last ~5) years using kWave. </p> <p>Thank you, thank you, thank you. Bless your heart for making this freely available to the public! </p> https://www.k-wave.org/forum/topic/size-of-kwavearray#post-9163 Mon, 09 Dec 2024 19:53:37 +0000 ‪Alon 9163@https://www.k-wave.org/forum/ <p>Hello,</p> <p>I've been experimenting with kWaveArray and noticed an issue when converting a karray to a binary mask. After conversion, the size of the elements seems to differ from the expected size, likely due to interpolation of off-grid points.</p> <p>For example, if I create a rectangular element with specific width and length, the binary mask generated by getArrayBinaryMask shows a size that is (much) larger than expected. Specifically, when I measure (in grid points) the length (say, along the x-axis), the result exceeds the intended size of Lx / dx (by more than 1-2 extra grid points).</p> <p>I understand that parameters like 'BLITolerance' can influence this behavior, but I'm wondering how I can ensure the size of the binary mask matches the intended dimensions exactly.</p> <p>Thanks for your help! </p> https://www.k-wave.org/forum/topic/running-thermal-simulations-on-a-gpu#post-9160 Thu, 21 Nov 2024 18:29:58 +0000 newmam 9160@https://www.k-wave.org/forum/ <p>Hi, I've been struggling to get the kWaveDiffusion function to run with "Datacast" set to "gpuarray-single". The github says support for this was added last year but I get an error when I attempt it to take a timestep. Is there a way to get around this? </p> https://www.k-wave.org/forum/topic/applying-multiple-frequencies-to-kwavetransducer#post-9150 Tue, 08 Oct 2024 06:42:34 +0000 donghunk 9150@https://www.k-wave.org/forum/ <p>Hello,</p> <p>I've been using the kWaveArray class to simulate different frequencies interacting with each other for vibro-acoustographic simulations.</p> <pre><code>% define multi-element kWaveArray source_signal(1, :) = ... % define 1st element signal source_signal(2, :) = ... % define 2nd element signal ... getDistributedSourceSignal(kgrid, source_signal)</code></pre> <p>I've been wondering if the same can be done with the kWaveTransducer class in order to make use of its B-mode functionality, but while having each element firing a different frequency. </p> <p>Currently, if I try to define a multi-row transducer.input_signal for a kWaveTransducer instance, it says</p> <pre><code>transducer.input_signal must be a one-dimensional array</code></pre> <p>Is there a workaround that exists? </p> https://www.k-wave.org/forum/topic/python-implementation-of-k-wave#post-9111 Thu, 11 Jul 2024 17:42:33 +0000 wsimson 9111@https://www.k-wave.org/forum/ <p>Hi All,</p> <p>as a follow-up, k-wave-python now supports MacOS out of the box in the latest release.</p> <p>Hope you find this helpful!</p> <p>Best,<br /> Walter </p> https://www.k-wave.org/forum/topic/modelling-nonlinear-sound-propagation-in-air#post-9098 Sat, 01 Jun 2024 14:38:45 +0000 DomeG 9098@https://www.k-wave.org/forum/ <p>... any update on this interesting topic? </p> https://www.k-wave.org/forum/topic/can-i-control-the-pixel-size-of-the-pressure-field-image#post-9097 Fri, 31 May 2024 05:28:56 +0000 Seoks 9097@https://www.k-wave.org/forum/ <p>I used kspaceFirstOrder2D and i extracted p_max data with 2D image.<br /> The grid size of simulation that i set is 551 by 551.<br /> But the p_max 2D image has 540 by 540 pixel size.<br /> So i wondered if there would be data loss during the image processing.<br /> Is there any way that i can control the output image pixel size? </p> https://www.k-wave.org/forum/topic/may-i-ask-if-there-is-any-code-similar-to-transcranial-ultrasound-simulation#post-9096 Fri, 31 May 2024 03:20:05 +0000 cyl 9096@https://www.k-wave.org/forum/ <p>May I ask if there is any code similar to transcranial ultrasound simulation that I would like to learn from </p> https://www.k-wave.org/forum/topic/focused-transducer-with-a-hyperbolic-reception-angle-like-ar-pam#post-9093 Wed, 15 May 2024 14:36:22 +0000 chengjinjin 9093@https://www.k-wave.org/forum/ <p>During the simulation, I found if I use an ark-shaped transducer to simulate a focused transducer, it can only realize geometry focus, but can not realize hyperbolic reception angle like AR-PAM. Do you know how to realize it with k-wave? </p> https://www.k-wave.org/forum/topic/after-getarraybinarymaskvoxelplot-visualize-the-overlap-problem#post-9071 Sat, 30 Mar 2024 03:42:49 +0000 HYH 9071@https://www.k-wave.org/forum/ <p>The added rectangular elements are 1 mm wide, 1 mm long, and 1.2 mm apart. However, after using the voxelPlot function to visualize, the red cubes are all overlapped, and after adjusting the interval to separate them, when the dx is 0.5mm, it is found that there are 20*20 red cubes in each rectangular array in voxelPlot. How can I solve this problem, my code is as follows。 Looking forward to your answer！</p> <pre><code>element_num_x = 8; element_num_y = 8; element_width = 1e-3; element_length = 1e-3; element_pitch = 1.2e-3; rotation = [0, 0, 0]; x_pos = (0:element_pitch:(element_num_x)*element_pitch); y_pos = (0:element_pitch:(element_num_x)*element_pitch); karray = kWaveArray(); for ind = 1:element_num_x for k = 1:element_num_y karray.addRectElement([x_pos(ind) ,y_pos(k) ,kgrid.z_vec(1)+ z_compensate], element_width, element_length, rotation); end end sensor.mask = karray.getArrayBinaryMask(kgrid); voxelPlot(sensor.mask, &#39;Color&#39;, [1 0 0], &#39;Transparency&#39;, 0.5);</code></pre> https://www.k-wave.org/forum/topic/calculating-2d-array-transducer-propagation-in-a-specific-z-depth-only#post-9066 Wed, 20 Mar 2024 20:54:33 +0000 Bradley Treeby 9066@https://www.k-wave.org/forum/ <p>Unfortunately not - full wave methods like k-Wave have to compute the propagation from the source to that depth. If your medium is homogeneous, you could instead use the Rayleigh integral or another toolbox like focus. </p> https://www.k-wave.org/forum/topic/one-sided-array-transducer#post-9061 Wed, 20 Mar 2024 20:40:33 +0000 Bradley Treeby 9061@https://www.k-wave.org/forum/ <p>Do you want to do this for a particular acoustic reason, or just for visualisation? The actual transducer housing etc is not modelled, so it’s not straightforward to do unfortunately. </p> https://www.k-wave.org/forum/topic/what-acoustic-difference-is-too-large#post-9050 Sun, 17 Mar 2024 18:05:09 +0000 daga_pankaj 9050@https://www.k-wave.org/forum/ <p>I am trying to simulate where there the difference in density between foreground and background is ~ 8x. The speed of sound difference is around ~2x. So impedance difference is around 16x.</p> <p>Is that too large for spectral methods like K-wave? </p>