http://www.k-wave.org/forum/topic/frequency-spectrum-of-point-source-in-3d Support for the k-Wave MATLAB toolbox en-US Wed, 13 May 2026 01:48:27 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/frequency-spectrum-of-point-source-in-3d#post-7415 Sat, 18 Apr 2020 11:07:01 +0000 Bradley Treeby 7415@http://www.k-wave.org/forum/ <p>Hi punnoj2,</p> <p>The frequency response of a point source in 1D emitting a delta function will be the Fourier transform of the delta function - so perfectly broadband. For a point source in 3D it will be the FT of the derivative of the delta function, so multiplied by I\omega, so you should see a ramp-type frequency response.</p> <p>Brad. </p> http://www.k-wave.org/forum/topic/frequency-spectrum-of-point-source-in-3d#post-7368 Sun, 05 Apr 2020 00:00:13 +0000 punnoj2 7368@http://www.k-wave.org/forum/ <p>Hello,</p> <p>I am trying to simulate a point source in 3D. I started by simply modifying the file example_na_source_smoothing.m to go from a 1D to a 3D model. After only changing the grid for source and sensor positions from 1D to 3D, I'm finding that I don't get the flat frequency spectrum that is demonstrated in the 1D example. Is there something that I'm doing wrong?</p> <p>Code Excerpt:<br /> % create the computational grid<br /> Nx = 100; % number of grid points<br /> dx = 0.05e-3; % grid point spacing [m]<br /> Ny = 100;<br /> Nz = 100;<br /> kgrid = kWaveGrid(Nx, dx, Ny, dx, Nz, dx);</p> <p>% define the properties of the propagation medium<br /> medium.sound_speed = 1500; % [m/s]</p> <p>% define time array<br /> dt = 2e-9; % [s]<br /> t_end = 4.26e-6; % [s]<br /> Nt = round(t_end / dt) + 1;<br /> kgrid.setTime(Nt, dt);</p> <p>% define the position of the source and sensor<br /> source_sensor_dist = 1e-6; % [s]<br /> source_pos = [Nx/2,Ny/2,Nz/2];<br /> sensor_pos = source_pos + [0,0,round(medium.sound_speed * source_sensor_dist / dx)];</p> <p>% create a binary sensor mask<br /> sensor.mask = zeros(Nx,Ny,Nz);<br /> sensor.mask(sensor_pos(1),sensor_pos(2),sensor_pos(3)) = 1;</p> <p> sensor_data = kspaceSecondOrder(kgrid, medium, source, sensor, 'Smooth', false, 'PlotSim', false);</p> <p> % calculate the amplitude spectrum of the recorded signal<br /> [f, func_as] = spect(sensor_data, 1/kgrid.dt);</p> <p> % plot the initial pressure distribution<br /> subplot(3, 3, (source_index * 3) - 2);<br /> plot_width = 10;<br /> stem(0:2*plot_width, source.p0(source_pos - plot_width:source_pos + plot_width), 'k');<br /> set(gca, 'XLim', [0, 2 * plot_width], 'YLim', [-0.1, 1.1]);<br /> xlabel('Sample');<br /> ylabel('Amplitude [au]');<br /> title([window_type ': Spatial Source Shape']);</p> <p> % plot the recorded time signals<br /> subplot(3, 3, (source_index * 3) - 1);<br /> [~, scale, prefix] = scaleSI(kgrid.t_array(end));<br /> plot(kgrid.t_array * scale, sensor_data(1, :), 'k-');<br /> axis tight;<br /> set(gca, 'YLim', [-0.15, 0.55], 'XLim', [1, 3]);<br /> ylabel('Amplitude [au]');<br /> xlabel(['Time [' prefix 's]']);<br /> title([window_type ': Recorded Time Pulse']);</p> <p> % plot the amplitude spectrum of the recorded signal<br /> subplot(3, 3, (source_index * 3));<br /> [~, scale, prefix] = scaleSI(f(end));<br /> func_as(1) = 2 * func_as(1);<br /> plot(f * scale, func_as ./ max(func_as(:)), 'k-');<br /> set(gca, 'XLim', [0, 20], 'YLim', [0, 1.05]);<br /> xlabel(['Frequency [' prefix 'Hz]']);<br /> ylabel('Relative Amplitude Spectrum');<br /> title([window_type ': Frequency Response']); </p>