Package gov.nih.mipav.model.algorithms

Class InverseLaplace

java.lang.Object
gov.nih.mipav.model.algorithms.InverseLaplace
Direct Known Subclasses:
AlgorithmFRAP.FitExpModel, InverseLaplaceTest.FitdeHoog1, InverseLaplaceTest.FitdeHoog10, InverseLaplaceTest.FitdeHoog11, InverseLaplaceTest.FitdeHoog12, InverseLaplaceTest.FitdeHoog13, InverseLaplaceTest.FitdeHoog14, InverseLaplaceTest.FitdeHoog15, InverseLaplaceTest.FitdeHoog16, InverseLaplaceTest.FitdeHoog2, InverseLaplaceTest.FitdeHoog3, InverseLaplaceTest.FitdeHoog4, InverseLaplaceTest.FitdeHoog5, InverseLaplaceTest.FitdeHoog6, InverseLaplaceTest.FitdeHoog7, InverseLaplaceTest.FitdeHoog8, InverseLaplaceTest.FitdeHoog9
public abstract class InverseLaplace extends Object
This is a port of the MATLAB INVLAP.M, a numerical inverse Laplace transform using the de Hoog algorithm, copyright by Karl Hollenbeck on November 22, 1996, Department of Hydrodynamics and Water Resources, Technical University of Denmark, DK-2800 Lyngby email:karl@isv16.isva.dtu.dk Downloaded from MATLAB central in ECP1 Software.zip.

The algorithm used is deHoog et al's quotient difference method with accelerated convergence for the continued fraction expansion published in "An improved method for numerical inversion of Laplace transforms" by F. R. de Hoog, J. H. Knight, and A. N. Stokes, SIAM Journal on Scientific and Statistical Computing, Vol. 3, No. 3, September, 1982, pp. 357-366.

Hollenbeck algorithm modification: The time vector is split in segments of equal magnitude which are inverted individually. This gives better overall accuracy.

From the original MATLAB source code: 

 Copyright: Karl Hollenbeck
 Department of Hydrodynamics and Water Resources
 Technical University of Denmark, DK-2800 Lyngby
 email: karl@isv16.isva.dtu.dk
 
 22 Nov 1996, MATLAB 5 version 27 Jun 1997 updated 1 Oct 1998
 
 IF YOU PUBLISH WORK BENEFITING FROM THIS M-FILE, PLEASE CITE IT AS:
 Hollenbeck, K. J. (1998) INVLAP.M: A matlab function for numerical 
 inversion of Laplace transforms by the de Hoog algorithm, 
 http://www.isva.dtu.dk/staff/karl/invlap.htm

  • Field Summary

    Fields
    Modifier and Type
    Field
    Description
    private double
    largestPole
    Don't include a time == 0.
    private double[]
    logT
    DOCUMENT ME!
    private double[]
    time
    DOCUMENT ME!
    private double[]
    timeFunction
    resulting vector of real-space values.
    private double
    tol
    numerical tolerance of approaching pole (default 1.0e-9).
    private int
    tPts
    DOCUMENT ME!

  • Constructor Summary

    Constructors
    Constructor
    Description
    InverseLaplace(double[] time)
    Constructor for InverseLaplace.
    InverseLaplace(double[] time, double largestPole, double tol)
    Constructor for InverseLaplace.

  • Method Summary

    Modifier and Type
    Method
    Description
    void
    driver()
    driver.
    abstract double[][]
    fitToLaplace(double realS, double[] imagS)
    Returns 'a' coefficients in power series where the second index is 0 for the real part and 1 for the imaginary part.
    double[]
    getTimeFunction()
    getTimeFunction.

    Methods inherited from class java.lang.Object

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

  • Field Details

    • largestPole

      private double largestPole
      Don't include a time == 0. This will give a -inifinity when the log of the time is taken. largest pole of Laplace space function (default zero) Note that InverseLaplace may give incorrect results if the default value of largestPole is used. For example, the function exp(t) with the Laplace transform 1/(s-1) will give correct results if largestPole is set to 1.0, but will return incorrect negative values for time >= 1.0 if the largestPole default of 0.0 is used.

    • logT

      private double[] logT
      DOCUMENT ME!

    • time

      private double[] time
      DOCUMENT ME!

    • timeFunction

      private double[] timeFunction
      resulting vector of real-space values.

    • tol

      private double tol
      numerical tolerance of approaching pole (default 1.0e-9).

    • tPts

      private int tPts
      DOCUMENT ME!

  • Constructor Details

    • InverseLaplace

      public InverseLaplace(double[] time)
      Constructor for InverseLaplace.
      Parameters:
      time - DOCUMENT ME!

    • InverseLaplace

      public InverseLaplace(double[] time, double largestPole, double tol)
      Constructor for InverseLaplace.
      Parameters:
      time - DOCUMENT ME!
      largestPole - DOCUMENT ME!
      tol - DOCUMENT ME!

  • Method Details

    • fitToLaplace

      public abstract double[][] fitToLaplace(double realS, double[] imagS)
      Returns 'a' coefficients in power series where the second index is 0 for the real part and 1 for the imaginary part.
      Parameters:
      realS - DOCUMENT ME!
      imagS - DOCUMENT ME!
      Returns:
      DOCUMENT ME!

    • driver

      public void driver()
      driver.

    • getTimeFunction

      public double[] getTimeFunction()
      getTimeFunction.
      Returns:
      timeFunction