Fast Fourier Transform


FFT

Inverse FFT

FFT-Object

This object calculates the fast fourier transform of a floating point vector. On output it returns the real and imaginary part in two different output vectors as well as the norm (square root of square sum of imaginary and real part) of the resulting complex vector. If a scale vector is given it is recalculated to a frequency scale.

For minimizing of unwanted frequencies the so called "Apodisation" can be enabled using the property dialog. The default value is "Subtr. mean". If this is choosen before doing the fourier transform the mean value of the input vector is calculated and subtracted from the input vector. The second choice is "Parabel": This means that after subtracting the mean value the whole input vector is multiplied with a negative parabolic function which is zero at both ends of the input data vector and 1 in the middle of the input data vector.

As a fast fourier transformation is used, the input vector normally must have the length of a power of 2. For the case the input vector doesn't have the right length, it is extended automatically to the next power of 2. Using the property dialog three methods can be specified where to get the newly created elements from:

Median
The newly created elements are set to the mean value of the input data.
Left
The newly created elements are set to the value of the first element of the input data vector.
Right
The newly created elements are set to the value of the last element of the input data vector.

The frequency spectrum resolution, which is given by the length of the input vector, can be increased by powers of 2, which can be done via the property dialog. The increase is done by extending the input vector appropriately.

Inverse FFT-Object

The inverse FFT performes an inverse fourier transform of the given real and imaginary data vectors. This can for example be used together with the forward transform for filtering in the frequency domain. The lengths of input vectors must be a power of 2, as they are if the output of the FFT object is used. FFT and following inverse FFT leads to the original input vector if apodisation is switched of in the FFT object.


The Vimms User Manual