k-Wave User Forum » Topic: 2D simple case-confused about PML

k-Wave User Forum » Topic: 2D simple case-confused about PML http://www.k-wave.org/forum/topic/2d-simple-case-confused-about-pml Support for the k-Wave MATLAB toolbox en-US Wed, 13 May 2026 03:10:36 +0000 http://bbpress.org/?v=1.0.2 <![CDATA[Search]]> q http://www.k-wave.org/forum/search.php Bradley Treeby on "2D simple case-confused about PML" http://www.k-wave.org/forum/topic/2d-simple-case-confused-about-pml#post-7795 Wed, 09 Sep 2020 19:31:19 +0000 Bradley Treeby 7795@http://www.k-wave.org/forum/ Hi RuiLiu, If you turn off the PML, then the waves will wrap from one side of the domain to the other. This is due to the periodicity assumed by the FFT used to calculate spatial gradients in k-Wave. If you use the default PML parameters (pml_alpha = 2, pml_size = 20) and place your source position outside the PML, e.g., at grid point 21 instead of 2, then the simulation runs as expected. Note, your source frequency is above the maximum frequency supported by the grid. If calling <code>filterTimeSeries</code> you might also want to pad the input signal with some zeros to allow the filter some room to work. Brad RuiLiu on "2D simple case-confused about PML" http://www.k-wave.org/forum/topic/2d-simple-case-confused-about-pml#post-7762 Fri, 21 Aug 2020 19:33:06 +0000 RuiLiu 7762@http://www.k-wave.org/forum/ Hi, Here is my code to put one source and two sensors in a line of the medium. The question is why my source wave always appear in my opposite boundary no matter which location I put for source. I know we should put the source out of the PML, but even when I put PML as 0, it still run to opposite boundary first. Could anyone help me to understand the prinple of the model I hope the source running from left to right and not suddenly appeared in the opposite way. Thanks. Here is my code: clc clear close all % ========================================================================= % SIMULATION % ========================================================================= % create the computational grid Nx = 60; % number of grid points= in the x (row) direction Ny = 120; % number of grid points in the y (column) direction dx = 1e-3; % grid point spacing in the x direction [m] 1mm dy = 1e-3; % grid point spacing in the y direction [m] kgrid = makeGrid(Nx, dx, Ny, dy); % define the properties of the propagation medium medium.sound_speed = 1500; % [m/s] % set the CFL cfl = 0.05; % set end time t_end = 50e-6; % end at 100s % create the time array [kgrid.t_array, dt] = makeTime(kgrid, medium.sound_speed, cfl, t_end); % define a single source point source.p_mask = zeros(Nx, Ny); source.p_mask(Nx/2, 2) = 1; % define a time varying sinusoidal source source_freq = 1e6; % [Hz] source_mag = 2; % [Pa] t_array = [0:0.1e-6:0.1e-4]; source.p = source_mag*sin(source_freq*t_array); % filter the source to remove high frequencies not supported by the grid source.p = filterTimeSeries(kgrid, medium, source.p); plot(t_array*1e6, source.p) % define a single sensor point sensor.mask = zeros(Nx, Ny); sensor.mask(Nx/2, Ny/2) = 1; sensor.mask(Nx/2, Ny-10) = 1; % define the acoustic parameters to record sensor.record = {'p'}; % define the properties of the PML to allow plane wave propagation pml_alpha = 0; % pml_size = [0, 0]; % set the input arguments input_args = {'PlotScale', 'auto', 'PMLSize', pml_size,... 'PMLAlpha', pml_alpha, 'PlotPML', true, 'PlotLayout', true}; % run the simulation sensor_data = kspaceFirstOrder2D(kgrid, medium, source, sensor, input_args{:}); %% % ========================================================================= % VISUALISATION % ========================================================================= % plot the sensor data figure; subplot(2, 1, 1) stackedPlot(kgrid.t_array*1e6, sensor_data.p(1,:)); ylim([-2 2]) xlabel('Time [\mus]'); subplot(2, 1, 2) stackedPlot(kgrid.t_array*1e6, sensor_data.p(2,:)); ylim([-2 2]) xlabel('Time [\mus]');