VIII Signal
processing and algorithms
The signal processing functions present in VEGA rely on the Frv analysis
library developped in VIRGO. A default version of the library is provided with
the VEGA sources but may be upgraded at any time.
The API of this library are not yet finalized but come very close to this
point.
This library uses the FFTW
(
http://www.fftw.org) library
for computing fft’s. The installation of FFTW is not difficult and
described in the installation documents, i.e. INSTALL file in the distribution
or
"Installation of ROOT"
A structure FrvRFFT that holds the result of an fft transform of a frame
vector FrVect is defined. The algorithm itself is build and called with
FrvRFFTFor (0,frvect) as for example in (suppose vect is an FrVect) :
vega[] fft = FrvRFFTFor (0,vect)
This will create a standard fft and apply it to the vector vect. On output,
the fft structure contains the result of the processing in the form of
FrVect’s that one can plot as usual :
vega[] gVM->Draw(fft->amplitude)
For the amplitude part of the fft. Once the fft object is not needed
anymore, the user should delete it with
vega[] FrvRFFTFree(fft)
Look at the example scrollspectrum.C in the tutorials.
More information will be added as soon as the Frv library gets a standard
shape.
Damir BUSKULIC
Last update :19/11/2001;