#include <DigitalFilter.h>

Public Member Functions
virtual double *	filter (size_t n, const double *w) const

virtual void	filterOut (size_t n, const double w, double out) const =0

virtual void	filterGraph (TGraph *g, bool filterErrors=false) const

virtual void	filterReplace (size_t n, double *w) const

virtual void	impulse (size_t n, double *out, size_t delay=0) const

virtual double *	impulse (size_t n, size_t delay=0) const

TGraph *	impulseGraph (size_t n=101, double dt=1, size_t delay=50) const

virtual void	response (size_t n, TGraph amplitude_response, TGraph phase_response, TGraph **group_delay=0) const

virtual TGraph *	amplitudeResponse (size_t n=101) const

virtual TGraph *	phaseResponse (size_t n=101) const

virtual TGraph *	groupDelay (size_t n=101) const

double	avgDelay (double min_freq=0, double max_freq=1, int n=101) const

virtual TPad *	drawResponse (TPad *c=0, int n=101, int delay=50) const

virtual std::complex< double >	transfer (std::complex< double > z) const =0

Detailed Description

Base class for all digital filters. Any implementing filters must implement filterOut and transfer.

Definition at line 26 of file DigitalFilter.h.

Inheritance diagram for FFTtools::DigitalFilter:

[legend]

Member Function Documentation

double FFTtools::DigitalFilter::avgDelay	(	double	min_freq = `0`,
		double	max_freq = `1`,
		int	n = `101`
	)		const

Computes the average delay in samples between min_freq and max_freq (both in normalized frequency) by taking the weighted average of the group delay

Definition at line 943 of file DigitalFilter.cxx.

TPad * FFTtools::DigitalFilter::drawResponse	(	TPad *	c = `0`,
		int	n = `101`,
		int	delay = `50`
	)		const

virtual

Draws the amplitude, phase, group delay and impulse response into the pad. Returns the pad.

Definition at line 969 of file DigitalFilter.cxx.

double * FFTtools::DigitalFilter::filter	(	size_t	n,
		const double *	w
	)		const

virtual

Filter input, returning a newly allocated output. The caller is responsible for freeing the allocated memory.

Parameters

n	size of waveform
w	waveform to filter

Returns: newly allocated output

Definition at line 186 of file DigitalFilter.cxx.

void FFTtools::DigitalFilter::filterGraph	(	TGraph *	g,
		bool	filterErrors = `false`
	)		const

virtual

Filter graph (and y errors, if available) ``in place'' (with temporary allocation)

Parameters

g	graph to filter
filterErrors	true if g->GetEY() should be filtered (if it exists)

Definition at line 172 of file DigitalFilter.cxx.

virtual void FFTtools::DigitalFilter::filterOut	(	size_t	n,
		const double *	w,
		double *	out
	)		const

pure virtual

Filter input with previously allocated output.

Parameters

n	size of waveform to filter
w	input waveform
out	output waveform (assumed to be allocated)

Implemented in FFTtools::IIRFilter, FFTtools::FIRFilter, and FFTtools::DigitalFilterSeries.

void FFTtools::DigitalFilter::filterReplace	(	size_t	n,
		double *	w
	)		const

virtual

Filter ``in place'' (with temporary allocation)

Parameters

n	size of waveform
w	input waveform, output will also be written here

Definition at line 165 of file DigitalFilter.cxx.

void FFTtools::DigitalFilter::impulse	(	size_t	n,
		double *	out,
		size_t	delay = `0`
	)		const

virtual

Filter response to unit impulse. A delay may be given which is helpful for acausal filters. This works by applying filterOut on an input array with all zeroes and one one at position delay.

Parameters

n	number of samples wanted in impulse response
out	allocated memory where impulse response will be written
delay	the time (in samples) of the unit impulse

Definition at line 141 of file DigitalFilter.cxx.

double * FFTtools::DigitalFilter::impulse	(	size_t	n,
		size_t	delay = `0`
	)		const

virtual

Filter response to unit impulse that allocates new memory. A delay may be given which is helpful for acausal filters. The user is responsible for deleting te allocated memory.

Parameters

n	number of samples wanted in impulse response
delay	the time (in samples) of the unit impulse

Returns: a newly allocated array of the impulse response

Definition at line 158 of file DigitalFilter.cxx.

TGraph * FFTtools::DigitalFilter::impulseGraph	(	size_t	n = `101`,
		double	dt = `1`,
		size_t	delay = `50`
	)		const

Get impulse response as a newly allocated TGraph. Caller is responsbile for deletion.

Parameters

n	number of samples wanted in impulse response
dt	the sample rate of the graph
delay	the ttime (in samples) of the unit impulse

Returns: a TGraph of the impulse response

Definition at line 149 of file DigitalFilter.cxx.

void FFTtools::DigitalFilter::response	(	size_t	n,
		TGraph **	amplitude_response,
		TGraph **	phase_response,
		TGraph **	group_delay = `0`
	)		const

virtual

Compute amplitude response, phase response and group delay directly from the transfer function, with all the gory ensuing numerical problems. This method is provided becuase it is more efficient to calculate two (or three) of the responses at once than individually.

If any is 0, it is not computed. If any TGraph ** points to a TGraph * that points to 0, TGraph of the right size is allocated

Definition at line 21 of file DigitalFilter.cxx.

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

components/libRootFftwWrapper/include/DigitalFilter.h
components/libRootFftwWrapper/src/DigitalFilter.cxx

Public Member Functions

Detailed Description

Member Function Documentation