Fourier Transforms (baseband_tasks.fourier)

Introduction

The Fourier transform module contains classes that wrap various fast Fourier transform (FFT) packages, in particular numpy.fft and pyfftw.FFTW. The purpose of the module is to give the packages a common interface, and to allow individual transforms to be defined once, then re-used multiple times. This is especially useful for FFTW, which achieves its fast transforms through prior planning.

The module currently does not support Hermitian Fourier transforms - frequency-domain values are always treated as complex.

Using the Fourier Module

To make FFTs, easiest is to use the fft_maker factory. In our examples, we will use it with the numpy fft back-end:

>>> from baseband_tasks.fourier import fft_maker
>>> fft_maker.set('numpy')
<ScienceState fft_maker: NumpyFFTMaker()>

The set() method allows one to choose any of the FFT maker classes - e.g. NumpyFFTMaker or PyfftwFFTMaker. Package-level options, such as the flags to FFTW, can be passed as **kwargs.

To create a transform, we pass the time-dimension data array shape and dtype, transform direction (‘forward’ or ‘backward’), transform axis (if the data is multi-dimensional), normalization convention and sample rate:

>>> import numpy as np
>>> import astropy.units as u
>>> fft = fft_maker((1000,), 'float64', direction='forward', ortho=True,
...                 sample_rate=1.*u.kHz)

Here, we have chosen orthogonal normalization, which normalizes both the frequency and time-domain outputs by \(1 / n^{1/2}\), where \(n\) is the length of the time-domain array. We can now perform the transform by calling fft:

>>> y = np.sin(2. * np.pi * np.arange(1000))
>>> Y = fft(y)

Y is the Fourier transform of y. To obtain the inverse, we use the inverse method in fft:

>>> ifft = fft.inverse()
>>> y_copy = y.copy()
>>> yn = ifft(Y)
>>> np.allclose(y, y_copy)
True

Note that we compare to a copy of the input; if possible for a given Fourier implementation (e.g., in pyfftw` but not in `numpy`), the ``inverse implementation reuses input and output arrays of the forward transform to save memory, so at the end on would have yn is y.

To show information about the transform, we can simply print the instance:

>>> fft
<NumpyFFT direction=forward,
    axis=0, ortho=True, sample_rate=1.0 kHz
    Time domain: shape=(1000,), dtype=float64
    Frequency domain: shape=(501,), dtype=complex128>

To obtain the sample frequencies of Y, we can use the frequency property:

>>> fft.frequency[:10]  
<Quantity [0.   , 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008,
       0.009] kHz>

For multi-dimensional arrays, the sample frequencies are for the transformed axis.

Reference/API

baseband_tasks.fourier Package

Fourier transform module.

This module provides a standard interface to various FFT packages.

Routine listings

fft_maker : primary interface for creating FFT instances.

Implementation Notes

For each packages, there is a corresponding *FFTMaker class, which holds default information needed for creating an FFT instance. For instance, for PyfftwFFTMaker, this holds flags, threads, etc.

These *FFTMaker instances in turn can be used to create *FFT instances which are set up to do the FFT on data with a given shape, in a given direction, etc.

Classes

NumpyFFTMaker()	FFT factory class utilizing numpy.fft functions.
PyfftwFFTMaker([n_simd])	FFT factory class utilizing the pyfftw package.
fft_maker(shape, dtype, *[, direction, ...])	Create an FFT, with a settable default engine.

Class Inheritance Diagram

baseband_tasks.fourier.base Module

Base classes and tools for the fourier module.

Implementation Notes

The base classes provide common code for adding new FFT engines.

In particular, the FFTMakerBase class is subclassed to create *FFTMaker classes (where * stands for a package such as pyfftw or numpy). These classes are initialized with default information needed for creating an FFT instance. For instance, for PyfftwFFTMaker, this holds flags, threads, etc. Via the FFTMakerMeta meta class, all such maker classes are registered in the FFT_MAKER_CLASSES dict, keyed by a lower-case version of the name (with fftmaker removed).

These *FFTMaker instances in turn can be used, via their __call__ method, to create *FFT subclass instances, setting relevant attributes such as time_shape, time_dtype, axis, etc., and instantiating them for a given direction.

The *FFT classes themselves are most easily based on FFTBase, which defines properties for accessing the various attributes, a default __init__ method that allows on to create an instance for a given direction of the FFT, and a FFTBase.__call__() method for actually doing the FFT on data. The FFTBase class also defines convenience properties, such as FFTBase.frequency to get the sample frequencies, and FFTBase.inverse() for getting the FFT in the inverse direction.

Selection of a default FFT package is done via fft_maker, which stores a default *FFTMaker instance. It is based on astropy.utils.state.ScienceState, but adds a fft_maker.system_default factory that one can access by setting the state to None, as well as __new__ method which allows one to use the default *FFTMaker instance to create an *FFT instance.

Classes

FFTMakerBase()	Base class for all FFT factories.
FFTBase(direction)	Framework for single pre-defined FFT and its associated metadata.
fft_maker(shape, dtype, *[, direction, ...])	Create an FFT, with a settable default engine.
FFTMakerMeta(name, bases, dct)	Registry of FFT maker classes.

Variables

FFT_MAKER_CLASSES

Dict for storing FFT maker classes, indexed by their name or prefix.

Class Inheritance Diagram

baseband_tasks.fourier.numpy Module

FFT maker and class using the numpy.fft routines.

Classes

NumpyFFTBase([direction])	Single pre-defined FFT based on numpy.fft.
NumpyFFTMaker()	FFT factory class utilizing numpy.fft functions.

Class Inheritance Diagram

baseband_tasks.fourier.pyfftw Module

FFT maker and class using pyfftw routines.

Implementation Notes

The code for PyfftwFFTBase is relatively complex to ensure that the input and output arrays are re-used, even between the forward and backward transforms (if created using PyfftwFFTBase.inverse())

Classes

PyfftwFFTBase(direction)	Single pre-defined FFT based on pyfftw.FFTW.
PyfftwFFTMaker([n_simd])	FFT factory class utilizing the pyfftw package.

Class Inheritance Diagram