umontreal.iro.lecuyer.hups.F2wNetLFSR Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of ssj Show documentation

SSJ is a Java library for stochastic simulation, developed under the direction of Pierre L'Ecuyer, in the Département d'Informatique et de Recherche Opérationnelle (DIRO), at the Université de Montréal. It provides facilities for generating uniform and nonuniform random variates, computing different measures related to probability distributions, performing goodness-of-fit tests, applying quasi-Monte Carlo methods, collecting (elementary) statistics, and programming discrete-event simulations with both events and processes.

The newest version!



/*
 * Class:        F2wNetLFSR
 * Description:  digital net in base 2 starting from a linear feedback
                 shift register generator
 * Environment:  Java
 * Software:     SSJ 
 * Copyright (C) 2001  Pierre L'Ecuyer and Université de Montréal
 * Organization: DIRO, Université de Montréal
 * @author       
 * @since

 * SSJ is free software: you can redistribute it and/or modify it under
 * the terms of the GNU General Public License (GPL) as published by the
 * Free Software Foundation, either version 3 of the License, or
 * any later version.

 * SSJ is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.

 * A copy of the GNU General Public License is available at
   GPL licence site.
 */

package umontreal.iro.lecuyer.hups;  

import umontreal.iro.lecuyer.util.PrintfFormat;


/**
 * This class implements a digital net in base 2 starting from a
 * linear feedback shift register generator.  It is exactly the same
 * point set as the one defined in the class 
 * {@link umontreal.iro.lecuyer.hups.F2wCycleBasedLFSR F2wCycleBasedLFSR}. 
 *  See the description
 * of this class for more details on the way the point set is constructed.
 * 
 * 
 * Constructing a point set using this class, instead of using
 *  {@link umontreal.iro.lecuyer.hups.F2wCycleBasedLFSR F2wCycleBasedLFSR},
 * makes SSJ construct a digital net in base 2.  This is useful if one
 * wants to randomize using one of the randomizations included in the class
 * {@link umontreal.iro.lecuyer.hups.DigitalNet DigitalNet}.
 * 
 */
public class F2wNetLFSR extends DigitalNetBase2  
{
   private F2wStructure param;

    /* *
     * Constructs and stores the set of cycles for an LCG with
     *    modulus n and multiplier a.
     *   If pgcd(a, n) = 1, this constructs a full-period LCG which has two
     *   cycles, one containing 0 and one, the LCG period.
     *
     * @param n required number of points and modulo of the LCG
     *
     *    @param a generator a of the LCG
     *
     *
     */



   /**
    * Constructs a point set with 2^rw points.  See the description of the class
    * {@link umontreal.iro.lecuyer.hups.F2wStructure F2wStructure} 
    * for the meaning of the parameters.
    * 
    */
   public F2wNetLFSR (int w, int r, int modQ, int step, int nbcoeff,
                      int coeff[], int nocoeff[], int dim)  
   {
      param = new F2wStructure (w, r, modQ, step, nbcoeff, coeff, nocoeff);
      param.initParamLFSR ();
      initNet (r, w, dim);
   }



   /**
    * Constructs a point set after reading its parameters from
    *    file filename; the parameters are located at line numbered no
    *    of filename. The available files are listed in the description of class
    * {@link umontreal.iro.lecuyer.hups.F2wStructure F2wStructure}.
    * 
    */
   public F2wNetLFSR (String filename, int no, int dim)  
   {
      param = new F2wStructure (filename, no);
      param.initParamLFSR ();
      initNet (param.r, param.w, dim);
   }



   public String toString ()
   {
       String s = "F2wNetLFSR:" + PrintfFormat.NEWLINE;
       return s + param.toString ();
   }


   private void initNet (int r, int w, int dim)
   {
      numCols = r * w;
      numRows = 31;
      outDigits = 31;
      numPoints = (1 << numCols);
      this.dim = dim;
      normFactor = 1.0 / (1L << 31);
      genMat = new int[dim * numCols];

      for (int j = 0; j < numCols; j++) {
         param.state = 1 << (r * w - 1 - j);
         param.initF2wLFSR ();
         genMat[j] = param.output;
         for (int i = 1; i < dim; i++) {
            param.F2wLFSR ();
            genMat[i * numCols + j] = param.output;
         }
      }
   }
}