Spectral kurtosis from signal or spectrogram

returns the spectral
kurtosis of vector `sk`

= pkurtosis(`x`

)`x`

as the vector
`sk`

. `pkurtosis`

uses normalized
frequency (evenly spaced frequency vector spanning [0 π]) to compute the time
values. `pkurtosis`

computes the spectrogram of
`x`

using `pspectrum`

with default
window size (time resolution in samples), and 80% window overlap.

returns the spectral kurtosis using the spectrogram or power spectrogram
`sk`

= pkurtosis(`s`

,`sampx`

,`f`

,`window`

)`s`

, along with:

Sample rate or time,

`sampx`

, of the original time-series signal that was transformed to produce`s`

Spectrogram frequency vector

`f`

Spectrogram time resolution

`window`

Use this syntax when you want to customize the options for
`pspectrum`

, rather than accept the default
`pspectrum`

options that `pkurtosis`

applies. You can specify `sampx`

as empty to default to
normalized frequency. Although `window`

is optional for
previous syntaxes, you must supply a value for `window`

when
using this syntax.

`[___,`

returns
the spectral kurtosis threshold `thresh`

] = pkurtosis(___,`'ConfidenceLevel'`

,`p`

) `thresh`

using the confidence
level `p`

. `thresh`

represents the range
within which the spectral kurtosis indicates a Gaussian stationary signal, at
the optional confidence level `p`

that you either specify or
accept as default. Specifying `p`

allows you to tune the
sensitivity of the spectral kurtosis `thresh`

results to
behavior that is non-Gaussian or nonstationary. You can use the
`thresh`

output argument with any of the input arguments
in previous syntaxes. You can also set the confidence level in previous
syntaxes, but it has no effect unless you are returning or plotting
`thresh`

.

`pkurtosis(___)`

plots the spectral kurtosis,
along with confidence level and thresholds, without returning any data. You can
use this syntax with any of the input arguments in previous syntaxes.

[1] Antoni, J., and R. B. Randall.
"The Spectral Kurtosis: Application to the Vibratory Surveillance and Diagnostics of
Rotating Machines." *Mechanical Systems and Signal Processing *.
Vol. 20, Issue 2, 2006, pp. 308–331.

[2] Antoni, J. "The Spectral
Kurtosis: A Useful Tool for Characterising Non-Stationary Signals."
*Mechanical Systems and Signal Processing*. Vol. 20, Issue 2,
2006, pp. 282–307.

`kurtogram`

| `pentropy`

| `pspectrum`

- Rolling Element Bearing Fault Diagnosis (Predictive Maintenance Toolbox)