qm-dsp  1.8
FFT Class Reference

#include <FFT.h>

List of all members.

Classes

class  D

Public Member Functions

 FFT (int nsamples)
 Construct an FFT object to carry out complex-to-complex transforms of size nsamples.
 ~FFT ()
void process (bool inverse, const double *realIn, const double *imagIn, double *realOut, double *imagOut)
 Carry out a forward or inverse transform (depending on the value of inverse) of size nsamples, where nsamples is the value provided to the constructor above.

Private Attributes

Dm_d

Detailed Description

Definition at line 12 of file FFT.h.


Constructor & Destructor Documentation

FFT::FFT ( int  nsamples)

Construct an FFT object to carry out complex-to-complex transforms of size nsamples.

nsamples does not have to be a power of two.

Definition at line 81 of file FFT.cpp.

FFT::~FFT ( )

Definition at line 86 of file FFT.cpp.

References m_d.


Member Function Documentation

void FFT::process ( bool  inverse,
const double *  realIn,
const double *  imagIn,
double *  realOut,
double *  imagOut 
)

Carry out a forward or inverse transform (depending on the value of inverse) of size nsamples, where nsamples is the value provided to the constructor above.

realIn and (where present) imagIn should contain nsamples each, and realOut and imagOut should point to enough space to receive nsamples each.

imagIn may be NULL if the signal is real, but the other pointers must be valid.

The inverse transform is scaled by 1/nsamples.

Definition at line 92 of file FFT.cpp.

References m_d, and FFT::D::process().

Referenced by ConstantQ::sparsekernel().


Member Data Documentation

D* FFT::m_d [private]

Definition at line 42 of file FFT.h.

Referenced by process(), and ~FFT().


The documentation for this class was generated from the following files: