k-Wave User Forum » Topic: Problem running linear transducer

k-Wave User Forum » Topic: Problem running linear transducer http://www.k-wave.org/forum/topic/problem-running-linear-transducer Support for the k-Wave MATLAB toolbox en-US Wed, 13 May 2026 03:45:24 +0000 http://bbpress.org/?v=1.0.2 <![CDATA[Search]]> q http://www.k-wave.org/forum/search.php Bradley Treeby on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4161 Sat, 30 Nov 2013 10:59:14 +0000 Bradley Treeby 4161@http://www.k-wave.org/forum/ It would seem that way. The functions need to be renamed from <code>.txt</code> to <code>.m</code> and placed in the root @GPUtype folder in the GPUmat toolbox. The GPUmat toolbox also needs to be added to the MATLAB path. I hope you manage to get it to work. Brad. SergioGwirc on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4160 Fri, 29 Nov 2013 20:25:59 +0000 SergioGwirc 4160@http://www.k-wave.org/forum/ Hi Brad It does not work. That means that I made a mistake when I have renamed functions fftn, ifftn, and bsxfun and placed in the folder /@GPUtype ? Sergio Bradley Treeby on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4157 Fri, 29 Nov 2013 16:52:25 +0000 Bradley Treeby 4157@http://www.k-wave.org/forum/ Hi Sergio, If you cast a 3D matrix to the GPU from the command line, do the <code>fftn</code> and <code>ifftn</code> functions work? i.e., something like: <pre><code>x = GPUsingle(rand(64, 64, 64)); y = fftn(x);</code></pre> Brad. SergioGwirc on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4156 Fri, 29 Nov 2013 16:16:27 +0000 SergioGwirc 4156@http://www.k-wave.org/forum/ Hi Brad Thank you for your answer. I renamed the functions and placed in /@GPUtype folder when I installed the GPUmat, but the errors remain. I followed a bit the program and the problem seems to be with bsxfun in kspaceFirstOrder3D (line 822). What I don't understand is that the program stops when bsxfun is used for the third time in line 822. Sergio Bradley Treeby on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4155 Fri, 29 Nov 2013 11:15:28 +0000 Bradley Treeby 4155@http://www.k-wave.org/forum/ Hi SergioGwirc If using the GPUmat toolbox with k-Wave, the functions <code>fftn</code>, <code>ifftn</code>, and <code>bsxfun</code> from the other/GPUmat" folder within the k-Wave Toolbox should be renamed to have <code>.m</code> extensions and placed in the /@GPUtype" folder within the GPUmat toolbox directory. Brad. SergioGwirc on "Problem running linear transducer" http://www.k-wave.org/forum/topic/problem-running-linear-transducer#post-4152 Thu, 28 Nov 2013 18:24:16 +0000 SergioGwirc 4152@http://www.k-wave.org/forum/ Dear Sir I intalled k-wave and trying to use with GPUmat and NVIDIA CUDA GTX 660. When I run the example "example_us_bmode_linear_transducer" after the GPUstart it gives an error and I not understand why. Could you help me? The command windows is as follows: Starting GPU - GPUmat version: 0.280 - Required CUDA version: 5.0 There is 1 device supporting CUDA CUDA Driver Version: 5.0 CUDA Runtime Version: 5.0 Device 0: "GeForce GTX 660" CUDA Capability Major revision number: 3 CUDA Capability Minor revision number: 0 Total amount of global memory: 2147483648 bytes - CUDA compute capability 3.0 ...done - Loading module EXAMPLES_CODEOPT - Loading module EXAMPLES_NUMERICS -> numerics30.cubin - Loading module NUMERICS -> numerics30.cubin - Loading module RAND >> example_us_bmode_linear_transducer k-Wave Transducer Properties transducer position: [1 22 42] transducer width: 11.8519mm (64 grid points) number of elements: 32 number of active elements: 32 (elements 1 to 32) element width: 370.3704um (2 grid points) element spacing (kerf): 0ym (0 grid points) element pitch: 370.3704um (2 grid points) element length: 4.4444mm (24 grid points) sound speed: 1540m/s focus distance: 20mm elevation focus distance: 19mm steering angle: 0 degrees steering angle max: auto degrees Computing scan line 1 of 96 Running k-Wave simulation... start time: 28-Nov-2013 14:08:11 reference sound speed: 1600m/s prepending transducer.input_signal with 17 leading zeros appending transducer.input_signal with 17 trailing zeros dt: 36.075ns, t_end: 57.1429us, time steps: 1585 input grid size: 216 by 108 by 108 grid points (40 by 20 by 20mm) maximum supported frequency: 3.78MHz expanding computational grid... computational grid size: 256 by 128 by 128 grid points casting variables to kWaveGPUsingle type... precomputation completed in 4.5368s starting time loop... Undefined function 'ifftn' for input arguments of type 'GPUsingle'. Error in kspaceFirstOrder3D (line 822) ux_sgx = bsxfun(@times, pml_x_sgx, ... Error in example_us_bmode_linear_transducer (line 240) sensor_data = kspaceFirstOrder3D(kgrid, medium, transducer, transducer, input_args{:});