help-gsl
[Top][All Lists]

## [Help-gsl] FFT

 From: Klaus Huthmacher Subject: [Help-gsl] FFT Date: Mon, 06 Jan 2014 15:30:33 +0100 User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130922 Icedove/17.0.9

Dear fellows,


I have another question concerning the FFT of the GSL. I wanted to start low and tried to transform a sine signal.


Since I assume a data set I take into account real values and a length of any magnitude. Thus I end up with Section 16.7: Mixed - radix FFT routines for real data, am I right?


But to be honest I do not fully grasp between the two methods real_transform and halfcomplex_transform. The first leads to an expected \delta - peak while the second method leads to 'oscillating boundaries'.


Furthermore I am highly confused that the FFT functions only need the data in the sense of the 'values' of a function and not the corresponding x values.


I thought to be quiete familier with the fouriertransform by solving equations with paper and pencil but this GSL implementation is curious.

Can someone give me a little more insight?

In the appendix is a minimal example together with a makefile.

Kind regards,

Klaus Huthmacher.


main.cpp
Description: Text Data

Makefile
Description: Text document

huthmacher.vcf
Description: Vcard