#include <itkNormalVariateGenerator.h>

Inheritance diagram for itk::Statistics::NormalVariateGenerator:

Collaboration diagram for itk::Statistics::NormalVariateGenerator:

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef SmartPointer< Self >	Pointer
typedef NormalVariateGenerator	Self
typedef RandomVariateGeneratorBase	Superclass
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual const char *	GetNameOfClass () const
double	GetVariate ()
void	Initialize (int randomSeed)
Static Public Member Functions
static Pointer	New ()
Protected Member Functions
double	FastNorm (void)
	NormalVariateGenerator ()
virtual void	PrintSelf (std::ostream &os, Indent indent) const
virtual	~NormalVariateGenerator ()
Private Attributes
double	m_ActualRSD
double	m_Chic1
double	m_Chic2
int	m_ELEN
int	m_Gaussfaze
int *	m_Gausssave
double	m_GScale
int	m_Irs
int	m_LEN
int	m_LMASK
int	m_Lseed
int	m_Nslew
double	m_Rcons
double	m_Rscale
double	m_Scale
int	m_TLEN
int *	m_Vec1

Detailed Description

Normal random variate generator.

This generation method was initially developed and implemented by Martin Styner, University of North Carolina at Chapel Hill, and his colleagues.

You should run Initialize() function before call GetNormalVariate() function.

The followings are original comments.

Revision date 31 May 1996 This is a revised version of the algorithm decribed in

ACM Transactions on Mathematical Software, Vol 22, No 1 March 1996, pp 119-127.

It is somewhat faster, and uses less memory as the vector of variates is updated in-situ. It has passed all the same statistical tests as decribed in the TOMS article, plus others. Seems OK so far.

Works well with total pool of 1024 variates, and does not need two vectors of this size, so does less damage to cache. Has been tested for frequency of tail values which should occur once in a million. OK. Other usual tests OK. About 13 % faster than TOMS version.

FAST GENERATOR OF PSEUDO-RANDOM UNIT NORMAL VARIATES

C.S.Wallace, Monash University, 1994

To use this code, files needing to call the generator should include:

 #include "FastNorm.h"

and be linked with the maths library (-lm) FastNorm.h contains declaration of the initialization routine 'initnorm()', definition of a macro 'FastGauss' used to generate variates, and three variables used in the macro. Read below for calling conventions.

THIS CODE ASSUMES TWO'S-COMPLEMENT 32-BIT INTEGER ARITHMATIC. IT ALSO ASSUMES THE 'C' COMPILER COMPILES THE LEFT-SHIFT OPERATOR "<<" AS A LOGICAL SHIFT, DISCARDING THE SIGN DIGIT AND SHIFTING IN ZEROS ON THE RIGHT, SO " X << 1" IS EQUIVALENT TO " X+X ". IT ALSO ASSUMES THE RIGHT-SHIFT OPERATOR ">>" IS SIGN-PRESERVING, SO ( -2 >> 1) = -1, ( -1>>1) = -1.

A fast generator of pseudo-random variates from the unit Normal distribution. It keeps a pool of about 1000 variates, and generates new ones by picking 4 from the pool, rotating the 4-vector with these as its components, and replacing the old variates with the components of the rotated vector.

The program should initialize the generator by calling initnorm(seed) with seed a int integer seed value. Different seed values will give different sequences of Normals. Then, wherever the program needs a new Normal variate, it should use the macro FastGauss, e.g. in statements like: x = FastGauss; (Sets x to a random Normal value)

Definition at line 97 of file itkNormalVariateGenerator.h.

Member Typedef Documentation

typedef SmartPointer< const Self > itk::Statistics::NormalVariateGenerator::ConstPointer

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 105 of file itkNormalVariateGenerator.h.

typedef SmartPointer< Self > itk::Statistics::NormalVariateGenerator::Pointer

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 104 of file itkNormalVariateGenerator.h.

typedef NormalVariateGenerator itk::Statistics::NormalVariateGenerator::Self

Standard class typedefs.

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 102 of file itkNormalVariateGenerator.h.

typedef RandomVariateGeneratorBase itk::Statistics::NormalVariateGenerator::Superclass

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 103 of file itkNormalVariateGenerator.h.

Constructor & Destructor Documentation

itk::Statistics::NormalVariateGenerator::NormalVariateGenerator ( ) [protected]

virtual itk::Statistics::NormalVariateGenerator::~NormalVariateGenerator ( ) [protected, virtual]

Member Function Documentation

virtual::itk::LightObject::Pointer itk::Statistics::NormalVariateGenerator::CreateAnother ( void ) const [virtual]

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::Object.

double itk::Statistics::NormalVariateGenerator::FastNorm ( void ) [protected]

get a variate

virtual const char* itk::Statistics::NormalVariateGenerator::GetNameOfClass ( ) const [virtual]

Run-time type information (and related methods).

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

double itk::Statistics::NormalVariateGenerator::GetVariate ( ) [virtual]

get a variate using FastNorm function

Implements itk::Statistics::RandomVariateGeneratorBase.

void itk::Statistics::NormalVariateGenerator::Initialize ( int randomSeed )

generate random number table

static Pointer itk::Statistics::NormalVariateGenerator::New ( ) [static]

Method for creation through the object factory.

Reimplemented from itk::Object.

virtual void itk::Statistics::NormalVariateGenerator::PrintSelf	(	std::ostream &	os,
		Indent	indent
	)		const `[protected, virtual]`

Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.

Reimplemented from itk::Object.