Module tagtraum.jipesfft

Package com.tagtraum.jipesfft

Class FFT

java.lang.Object

com.tagtraum.jipesfft.AbstractFFT

com.tagtraum.jipesfft.FFT

All Implemented Interfaces:

AutoCloseable

public class FFT
extends AbstractFFT

Native implementation of FFT for a specific number of samples. Uses PureJavaFFT as fallback for some cases. An instance may be re-used for efficiency, but is not necessarily thread-safe.

Author:

Hendrik Schreiber

Field Summary

Fields inherited from class com.tagtraum.jipesfft.AbstractFFT

FREQUENCY, IMAGINARY, REAL

Constructor Summary

Constructors

Constructor	Description
FFT(int numberOfSamples)	Constructor for a given number of samples.

Method Summary

Modifier and Type	Method	Description
void	close()
float[][]	inverseTransform(float[] real, float[] imaginary)	Perform inverse transform.
float[][]	transform(float[] real)	Transform for real numbers.
float[][]	transform(float[] real, float[] imaginary)	Transform for complex numbers.

Methods inherited from class com.tagtraum.jipesfft.AbstractFFT

getFrequencies, getFrequencyForBin, getNumberOfSamples, toString

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

FFT

public FFT(int numberOfSamples)

Constructor for a given number of samples.

Parameters:

numberOfSamples - number of samples you intend to transform, must be a power of two

Method Detail

inverseTransform

public float[][] inverseTransform(float[] real,
                                  float[] imaginary)
                           throws UnsupportedOperationException

Description copied from class: AbstractFFT

Perform inverse transform. For very large number of samples (>32768) the Java version is used.

Specified by:

inverseTransform in class AbstractFFT

Parameters:

real - real part

imaginary - imaginary part

Returns:

two-dimensional array for the real and the imaginary parts

Throws:

UnsupportedOperationException

transform

public float[][] transform(float[] real)
                    throws UnsupportedOperationException

Description copied from class: AbstractFFT

Transform for real numbers. For very large number of samples (>32768) the Java version is used.

Specified by:

transform in class AbstractFFT

Parameters:

real - samples

Returns:

three-dimensional array, consisting of the real part, the imaginary, and the frequencies

Throws:

UnsupportedOperationException

transform

public float[][] transform(float[] real,
                           float[] imaginary)
                    throws UnsupportedOperationException

Description copied from class: AbstractFFT

Transform for complex numbers. Note that the current implementation simply delegates to the pure Java version.

Specified by:

transform in class AbstractFFT

Parameters:

real - real part

imaginary - imaginary part

Returns:

three-dimensional array, consisting of the real part, the imaginary, and the frequencies

Throws:

UnsupportedOperationException

close

public void close()