FFT routines for GWpy¶

The Fourier transform (and the associated digital FFT algorithm) is the most common way of investigating the frequency-domain content of a time-domain signal. GWpy provides wrappers of average spectrum methods from scipy.signal, lal, and pycbc.psd to simplify calculating a FrequencySeries from a TimeSeries.

The gwpy.signal.fft sub-package provides the following FFT averaging methods:

Method name	Function
`'pycbc_welch'`	`gwpy.signal.fft.pycbc.welch`
`'pycbc_bartlett'`	`gwpy.signal.fft.pycbc.bartlett`
`'pycbc_median'`	`gwpy.signal.fft.pycbc.median`
`'pycbc_median_mean'`	`gwpy.signal.fft.pycbc.median_mean`
`'lal_welch'`	`gwpy.signal.fft.lal.welch`
`'lal_bartlett'`	`gwpy.signal.fft.lal.bartlett`
`'lal_median'`	`gwpy.signal.fft.lal.median`
`'lal_median_mean'`	`gwpy.signal.fft.lal.median_mean`
`'scipy_welch'`	`gwpy.signal.fft.scipy.welch`
`'scipy_bartlett'`	`gwpy.signal.fft.scipy.bartlett`

Each of these can be specified by passing the function name as the method keyword argument to any of the relevant TimeSeries instance methods, e.g:

>>> ts = TimeSeries(...)
>>> psd = ts.psd(..., method='pycbc_welch', ...)

Additionally, short versions of the following registrations map to the first registered instance of that name in the registry:

Method name	Function
`'welch'`	`gwpy.signal.fft.pycbc.welch`
`'bartlett'`	`gwpy.signal.fft.pycbc.bartlett`
`'median'`	`gwpy.signal.fft.pycbc.median`
`'median_mean'`	`gwpy.signal.fft.pycbc.median_mean`

Note

The simple names (e.g. 'welch') are mapped at runtime, meaning if you don’t have pycbc on your system, the registrations will map to lal or scipy instead. These are provided only for convenience, so if you want to be able to reproduce your results on any system, use the full API-specific name when using TimeSeries.psd() and friends.

Note

You can check which FFT library will be used via the gwpy.signal.fft.get_default_fft_api() method, e.g:

>>> from gwpy.signal.fft import get_default_fft_api
>>> get_default_fft_api()
'pycbc.psd'

Notes¶

PyCBC methods¶

Each of the pycbc_* FFT methods accepts the following keyword arguments: o

Keyword	Default	Type	Description
`scheme`	`None`	`pycbc.scheme.Scheme`	processing scheme in which to execute FFT

LAL methods¶

Each of the lal_* FFT methods accepts the following keyword arguments:

Keyword	Default	Type	Description
`window`	`None`	`tuple`, `str`	time-doman window definition see `gwpy.signal.fft.lal.generate_window()`
`plan`	`None`	`lal.REAL8FFTPlan` or similar	plan to use when processing FFT, see `gwpy.signal.fft.lal.generate_fft_plan()`