Skip to main content

FFT

class FFT​

FFT(n, is_inverse=False, is_real=True)

Creates an object to handle the Fast Fourier Transform.

  • n is the number of input elements that will be feed to the transform;
  • is_inverse if set to True the FFT class will perform the inverse Fast Furier Transform;
  • is_real if set to True the FFT will be optimized to transform an input array with only real elements. This will result in less resources consumption and faster computation but limit the FFT to use only real input.

The FFT class features __setitem__ to pass inputs and __getitem__ to get outputs. This allows to set inputs as fourier[i] = input_element and get outputs as output_element = fourier[i] with fourier as FFT object.

The FFT class also features __iter__ and __next__ to be iterable.

Returns a FFT object.

method fft​

fft()

Performs a Fast Fourier Transform of the n input data set. Before calling the fft method all the n inputs must be provided using the settable features of the FFT class. Once the Fast Fourier Transform is done, outputs can be accessed using the gettable feature of the FFT class.

method amplitude​

amplitude(i)

Get the amplitude of the i element of the output array.

  • i is the index of the output to compute the amplitude of.

Returns the amplitude (sqrt(|out|/n)) of the element.

method phase​

phase(i, threshold=0)

Get the phase of the i element of the output array. If the amplitude value of the i element is smaller than the threshold the phase is not computed and set to zero.

  • i is the index of the output to compute the phase of.
  • threshold is the amplitude threshold used to filter. Default value is 0.

*** note ***: the threshold is often necessary to prevent noise coming from the finite precision of floating point values.

Returns the phase (atan(out_imm/out_real)) of the element.