Main Page   Groups   Namespace List   Class Hierarchy   Alphabetical List   Compound List   File List   Namespace Members   Compound Members   File Members   Concepts

itk::Statistics::NormalVariateGenerator Class Reference

Normal random variate generator. More...

#include <itkNormalVariateGenerator.h>

Inheritance diagram for itk::Statistics::NormalVariateGenerator:

Inheritance graph
[legend]
Collaboration diagram for itk::Statistics::NormalVariateGenerator:

Collaboration graph
[legend]
List of all members.

Public Types

typedef NormalVariateGenerator Self
typedef RandomVariateGeneratorBase Superclass
typedef SmartPointer< SelfPointer
typedef SmartPointer< const
Self
ConstPointer

Public Member Functions

virtual const char * GetClassName () const
void Initialize (int randomSeed)
double GetVariate ()

Static Public Member Functions

Pointer New ()

Protected Member Functions

 NormalVariateGenerator ()
virtual ~NormalVariateGenerator ()
virtual void PrintSelf (std::ostream &os, Indent indent) const
double FastNorm (void)

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 the file "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 91 of file itkNormalVariateGenerator.h.


Member Typedef Documentation

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

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 99 of file itkNormalVariateGenerator.h.

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

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 98 of file itkNormalVariateGenerator.h.

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

Standard class typedefs.

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 96 of file itkNormalVariateGenerator.h.

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

Reimplemented from itk::Statistics::RandomVariateGeneratorBase.

Definition at line 97 of file itkNormalVariateGenerator.h.


Constructor & Destructor Documentation

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

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


Member Function Documentation

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

get a variate

virtual const char* itk::Statistics::NormalVariateGenerator::GetClassName  )  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

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.


The documentation for this class was generated from the following file:
Generated at Sat Mar 31 03:13:47 2007 for ITK by doxygen 1.3.8 written by Dimitri van Heesch, © 1997-2000