PowerSpectralDensity

class ansys.sound.core.spectral_processing.PowerSpectralDensity(input_signal, fft_size=2048, window_type='HANN', window_length=2048, overlap=0.25)

Power Spectral Density (PSD) class.

This class allows the calculation of Power Spectral Density (PSD) for a given signal, using Welch’s method.

See also

Stft

Examples

Compute and display the Power Spectral Density (PSD) of an acoustic signal (in Pa).

>>> from ansys.sound.core.spectral_processing import PowerSpectralDensity
>>> power_spectral_density = PowerSpectralDensity(input_signal=signal)
>>> power_spectral_density.process()
>>> psd_dB = power_spectral_density.get_PSD_dB(ref_value=2e-5)
>>> power_spectral_density.plot(display_in_dB=True, ref_value=2e-5)

Class instantiation takes the following parameters.

Parameters:

signalField

Input signal on which to compute the PSD.

fft_sizeint, default: 2048

Number of FFT points to use for the PSD estimate. Must be a power of 2.

window_typestr, default: ‘HANN’

Window type used for the PSD computation. Options are 'TRIANGULAR', 'BLACKMAN', 'HAMMING', 'HANN', 'GAUSS', 'FLATTOP', and 'RECTANGULAR'.

window_lengthint, default: 2048

Number of points of the window used for the PSD computation , by default 2048.

overlapfloat, default: 0.25

Overlap value between two successive segments where the FFT is computed. Values range from 0 to 1. For example, 0 means no overlap, and 0.5 means 50% overlap.

Methods

PowerSpectralDensity.get_PSD_dB([ref_value])	Get the PSD in dB/Hz, as a Field.
PowerSpectralDensity.get_PSD_dB_as_nparray([...])	Get the PSD in dB/Hz as a NumPy array.
PowerSpectralDensity.get_PSD_squared_linear()	Get the PSD in squared linear unit per Hz.
PowerSpectralDensity.get_PSD_squared_linear_as_nparray()	Get the PSD in squared linear unit per Hz, as NumPy arrays.
PowerSpectralDensity.get_frequencies()	Get the frequencies associated with the PSD.
PowerSpectralDensity.get_output()	Get the PSD data as a DPF field.
PowerSpectralDensity.get_output_as_nparray()	Get the PSD data as NumPy arrays.
PowerSpectralDensity.plot([display_in_dB, ...])	Plot the PSD.
PowerSpectralDensity.process()	Calculate the PSD.

Attributes

PowerSpectralDensity.fft_size	Number of FFT points.
PowerSpectralDensity.input_signal	Input signal.
PowerSpectralDensity.overlap	Window overlap in %.
PowerSpectralDensity.window_length	Number of window points.
PowerSpectralDensity.window_type	Window type.