trunk/SHAPES/FST_semi_memo.c

/***************************************************************************
  **************************************************************************
  
                           S2kit 1.0

          A lite version of Spherical Harmonic Transform Kit

   Peter Kostelec, Dan Rockmore
   {geelong,rockmore}@cs.dartmouth.edu
  
   Contact: Peter Kostelec
            geelong@cs.dartmouth.edu
  
   Copyright 2004 Peter Kostelec, Dan Rockmore

   This file is part of S2kit.

   S2kit is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.

   S2kit 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.

   You should have received a copy of the GNU General Public License
   along with S2kit; if not, write to the Free Software
   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

   See the accompanying LICENSE file for details.
  
  ************************************************************************
  ************************************************************************/


/********************************************************************

  FST_semi_memo.c - routines to perform convolutions on the
  2-sphere using a combination of semi-naive and naive algorithms.

  ASSUMES THAT ALL PRECOMPUTED-DATA IS IN MEMORY, AND NOT TO BE
  READ FROM THE DISK.

  The primary functions in this package are

  1) FST_semi_memo() - computes the spherical harmonic expansion.
  2) InvFST_semi_memo() - computes the inverse spherical harmonic transform.
  3) FZT_semi_memo() - computes the zonal harmonic transform.
  4) TransMult() - Multiplies harmonic coefficients using Driscoll-Healy
                    result.  Dual of convolution in "time" domain.
  5) Conv2Sphere_semi_memo() - Convolves two functins defined on the 2-sphere,
                          using seminaive transform

  and one utility function:

  1) seanindex(): Given bandwidth bw, seanindex(m,l,bw) will give the
     position of the coefficient f-hat(m,l) in the one-row array

  For descriptions on calling these functions, see the documentation
  preceding each function.

*/

#include <math.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "fftw3.h"

#include "makeweights.h"
#include "cospmls.h"
#include "primitive.h"
#include "naive_synthesis.h"
#include "seminaive.h"
#include "FST_semi_memo.h"


/************************************************************************/


/*****************************************************************

  Given bandwidth bw, seanindex(m,l,bw) will give the position of the
  coefficient f-hat(m,l) in the one-row array that Sean stores the spherical
  coefficients. This is needed to help preserve the symmetry that the
  coefficients have: (l = degree, m = order, and abs(m) <= l)

  f-hat(l,-m) = (-1)^m * conjugate( f-hat(l,m) )

  Thanks for your help Mark!

  ******************************************************************/


int seanindex(int m,
              int l,
              int bw)
{     
  int bigL;

  bigL = bw - 1;

  if( m >= 0 )
    return( m * ( bigL + 1 ) - ( ( m * (m - 1) ) /2 ) + ( l - m ) );
  else
    return( ( ( bigL * ( bigL + 3 ) ) /2 ) + 1 +
            ( ( bigL + m ) * ( bigL + m + 1 ) / 2 ) + ( l - abs( m ) ) );
}


/************************************************************************/
/* performs a spherical harmonic transform using the semi-naive
   and naive algorithms

   bw -> bandwidth of problem
   size -> size = 2*bw -> dimension of input array (recall that
           sampling is done at twice the bandwidth)

   The inputs rdata and idata are expected to be pointers to
   size x size arrays. The array rdata contains the real parts
   of the function samples, and idata contains the imaginary
   parts.

   rcoeffs and icoeffs are expected to be pointers to bw x bw arrays,
   and will contain the harmonic coefficients in a "linearized" form.
   The array rcoeffs contains the real parts of the coefficients,
   and icoeffs contains the imaginary parts.

   spharmonic_pml_table should be a (double **) pointer to
   the result of a call to Spharmonic_Pml_Table.  Because this
   table is re-used in the inverse transform, and because for
   timing purposes the computation of the table is not included,
   it is passed in as an argument.  Also, at some point this
   code may be used as par of a series of convolutions, so
   reducing repetitive computation is prioritized.

   spharmonic_pml_table will be an array of (double *) pointers
   the array being of length TableSize(m,bw)

   workspace needs to be a double pointer to an array of size
   (8 * bw^2) + (7 * bw).

   cutoff -> what order to switch from semi-naive to naive
             algorithm.

   dataformat =0 -> samples are complex, =1 -> samples real

 
   Output Ordering of coeffs f(m,l) is
   f(0,0) f(0,1) f(0,2) ... f(0,bw-1)
          f(1,1) f(1,2) ... f(1,bw-1)
          etc.
                 f(bw-2,bw-2), f(bw-2,bw-1)
                               f(bw-1,bw-1)
                               f(-(bw-1),bw-1)
                 f(-(bw-2),bw-2) f(-(bw-2),bw-1)
          etc.
                  f(-2,2) ... f(-2,bw-1)
          f(-1,1) f(-1,2) ... f(-1,bw-1)

    
   This only requires an array of size (bw*bw).  If zero-padding
   is used to make the indexing nice, then you need a an
   (2bw-1) * bw array - but that is not done here.
   Because of the amount of space necessary for doing
   large transforms, it is important not to use any
   more than necessary.

*/

void FST_semi_memo(double *rdata, double *idata,
                   double *rcoeffs, double *icoeffs, 
                   int bw,
                   double **seminaive_naive_table,
                   double *workspace,
                   int dataformat,
                   int cutoff,
                   fftw_plan *dctPlan,
                   fftw_plan *fftPlan,
                   double *weights )
{
  int size, m, i, j;
  int dummy0, dummy1 ;
  double *rres, *ires;
  double *rdataptr, *idataptr;
  double *fltres, *scratchpad;
  double *eval_pts;
  double pow_one;
  double tmpA, tmpSize ;

  size = 2*bw ;
  tmpSize = 1./ ((double ) size);
  tmpA = sqrt( 2. * M_PI ) ;

  rres = workspace;  /* needs (size * size) = (4 * bw^2) */
  ires = rres + (size * size); /* needs (size * size) = (4 * bw^2) */ 
  fltres = ires + (size * size) ; /* needs bw  */
  eval_pts = fltres + bw; /* needs (2*bw)  */
  scratchpad = eval_pts + (2*bw); /* needs (4 * bw)  */
 
  /* total workspace is (8 * bw^2) + (7 * bw) */

  /* do the FFTs along phi */
  fftw_execute_split_dft( *fftPlan,
                          rdata, idata,
                          rres, ires ) ; 
  /*
    normalize

    note that I'm getting the sqrt(2pi) in there at
    this point ... to account for the fact that the spherical
    harmonics are of norm 1: I need to account for
    the fact that the associated Legendres are
    of norm 1
  */
  tmpSize *= tmpA ;
  for( j = 0 ; j < size*size ; j ++ )
    {
      rres[j] *= tmpSize  ;
      ires[j] *= tmpSize  ;
    }
  
  /* point to start of output data buffers */
  rdataptr = rcoeffs;
  idataptr = icoeffs;
  
  for (m=0; m<bw; m++) {
    /*
      fprintf(stderr,"m = %d\n",m);
    */
    
    /*** test to see if before cutoff or after ***/
    if (m < cutoff){
      
      /* do the real part */
      SemiNaiveReduced(rres+(m*size), 
                       bw, 
                       m, 
                       fltres,
                       scratchpad,
                       seminaive_naive_table[m],
                       weights,
                       dctPlan);
      
      /* now load real part of coefficients into output space */  
      memcpy(rdataptr, fltres, sizeof(double) * (bw - m));
      
      rdataptr += bw-m;
      
      /* do imaginary part */
      SemiNaiveReduced(ires+(m*size),
                       bw, 
                       m, 
                       fltres,
                       scratchpad,
                       seminaive_naive_table[m],
                       weights,
                       dctPlan);

      /* now load imaginary part of coefficients into output space */  
      memcpy(idataptr, fltres, sizeof(double) * (bw - m));
      
      idataptr += bw-m;
      
    }
    else{
      /* do real part */      
      Naive_AnalysisX(rres+(m*size),
                      bw,
                      m,
                      weights,
                      fltres,
                      seminaive_naive_table[m],
                      scratchpad );
      memcpy(rdataptr, fltres, sizeof(double) * (bw - m));
      rdataptr += bw-m;
      
      /* do imaginary part */
      Naive_AnalysisX(ires+(m*size),
                      bw,
                      m,
                      weights,
                      fltres,
                      seminaive_naive_table[m],
                      scratchpad );
      memcpy(idataptr, fltres, sizeof(double) * (bw - m));
      idataptr += bw-m;
    }
  }
  
  /*** now do upper coefficients ****/
  
  /* now if the data is real, we don't have to compute the
     coefficients whose order is less than 0, i.e. since
     the data is real, we know that
     f-hat(l,-m) = (-1)^m * conjugate(f-hat(l,m)),
     so use that to get the rest of the coefficients
     
     dataformat =0 -> samples are complex, =1 -> samples real
     
  */
  
  if( dataformat == 0 ){
    
    /* note that m is greater than bw here, but this is for
       purposes of indexing the input data arrays.  
       The "true" value of m as a parameter for Pml is
       size - m  */
    
    for (m=bw+1; m<size; m++) {
      /*
        fprintf(stderr,"m = %d\n",-(size-m));
      */
      
      if ( (size-m) < cutoff )
        {
          /* do real part */
          SemiNaiveReduced(rres+(m*size), 
                           bw, 
                           size-m, 
                           fltres,
                           scratchpad,
                           seminaive_naive_table[size-m],
                           weights,
                           dctPlan );
      
          /* now load real part of coefficients into output space */  
          if ((m % 2) != 0) {
            for (i=0; i<m-bw; i++)
              rdataptr[i] = -fltres[i];
          }
          else {
            memcpy(rdataptr, fltres, sizeof(double) * (m - bw));
          }
          rdataptr += m-bw;
      
          /* do imaginary part */
          SemiNaiveReduced(ires+(m*size), 
                           bw, 
                           size-m, 
                           fltres,
                           scratchpad,
                           seminaive_naive_table[size-m],
                           weights,
                           dctPlan);
      
          /* now load imag part of coefficients into output space */  
          if ((m % 2) != 0) {
            for (i=0; i<m-bw; i++)
              idataptr[i] = -fltres[i];
          }
          else {
            memcpy(idataptr, fltres, sizeof(double) * (m - bw));
          }
          idataptr += m-bw;
        }
      else
        {
          Naive_AnalysisX(rres+(m*size),
                          bw,
                          size-m,
                          weights,
                          fltres,
                          seminaive_naive_table[size-m],
                          scratchpad);

          /* now load real part of coefficients into output space */  
          if ((m % 2) != 0) {
            for (i=0; i<m-bw; i++)
              rdataptr[i] = -fltres[i];
          }
          else {
            memcpy(rdataptr, fltres, sizeof(double) * (m - bw));
          }
          rdataptr += m-bw;

          /* do imaginary part */
          Naive_AnalysisX(ires+(m*size),
                          bw,
                          size-m,
                          weights,
                          fltres,
                          seminaive_naive_table[size-m],
                          scratchpad);

          /* now load imag part of coefficients into output space */  
          if ((m % 2) != 0) {
            for (i=0; i<m-bw; i++)
              idataptr[i] = -fltres[i];
          }
          else {
            memcpy(idataptr, fltres, sizeof(double) * (m - bw));
          }
          idataptr += m-bw;

        }

    }
  }
  else        /**** if the data is real ****/
    {            
      pow_one = 1.0;
      for(i = 1; i < bw; i++){
        pow_one *= -1.0;
        for( j = i; j < bw; j++){
          /*
          SEANINDEXP(dummy0,i,j,bw);
          SEANINDEXN(dummy1,-i,j,bw);
          */

          dummy0 = seanindex(i, j, bw);
          dummy1 = seanindex(-i, j, bw);

          rcoeffs[dummy1] =
            pow_one * rcoeffs[dummy0];
          icoeffs[dummy1] =
            -pow_one * icoeffs[dummy0];
        }
      }
    }
  
}

/************************************************************************/
/* Inverse spherical harmonic transform.

   bw -> bandwidth of problem
   size = 2*bw

   Inputs rcoeffs and icoeffs are harmonic coefficients stored
   in (bw * bw) arrays in the order spec'ed above.

   rdata and idata are (size x size) arrays with the transformed result.

   transpose_spharmonic_pml_table should be the (double **)
   result of a call to Transpose_Spharmonic_Pml_Table()

   workspace is (8 * bw^2) + (10 * bw)

*/

/*      dataformat =0 -> samples are complex, =1 -> samples real */

void InvFST_semi_memo(double *rcoeffs, double *icoeffs, 
                      double *rdata, double *idata, 
                      int bw, 
                      double **transpose_seminaive_naive_table,
                      double *workspace,
                      int dataformat,
                      int cutoff,
                      fftw_plan *idctPlan,
                      fftw_plan *ifftPlan )
{
  int size, m, i, n;
  double *rdataptr, *idataptr;
  double *rfourdata, *ifourdata;
  double *rinvfltres, *iminvfltres, *scratchpad;
  double *sin_values, *eval_pts;
  double tmpA ;

  size = 2*bw ;
 
  rfourdata = workspace;                  /* needs (size * size) */
  ifourdata = rfourdata + (size * size);  /* needs (size * size) */
  rinvfltres = ifourdata + (size * size); /* needs (2 * bw) */
  iminvfltres = rinvfltres + (2 * bw);    /* needs (2 * bw) */
  sin_values = iminvfltres + (2 * bw);    /* needs (2 * bw) */
  eval_pts = sin_values + (2 * bw);       /* needs (2 * bw) */
  scratchpad = eval_pts + (2 * bw);       /* needs (2 * bw) */
  
  /* total workspace = (8 * bw^2) + (10 * bw) */

  /* load up the sin_values array */
  n = 2*bw;

  ArcCosEvalPts(n, eval_pts);
  for (i=0; i<n; i++)
    sin_values[i] = sin(eval_pts[i]);


  /* Now do all of the inverse Legendre transforms */
  rdataptr = rcoeffs;
  idataptr = icoeffs;

  for (m=0; m<bw; m++)
    {
      /*
        fprintf(stderr,"m = %d\n",m);
      */
    
      if(m < cutoff)
        {
          /* do real part first */ 
          InvSemiNaiveReduced(rdataptr,
                              bw,
                              m,
                              rinvfltres,
                              transpose_seminaive_naive_table[m],
                              sin_values,
                              scratchpad,
                              idctPlan );
      
          /* now do imaginary part */
      
          InvSemiNaiveReduced(idataptr,
                              bw,
                              m,
                              iminvfltres,
                              transpose_seminaive_naive_table[m],
                              sin_values,
                              scratchpad,
                              idctPlan);
          
          /* will store normal, then tranpose before doing inverse fft */
          memcpy(rfourdata+(m*size), rinvfltres, sizeof(double) * size);
          memcpy(ifourdata+(m*size), iminvfltres, sizeof(double) * size);
      
          /* move to next set of coeffs */
          rdataptr += bw-m;
          idataptr += bw-m;
      
        }
      else
        {

          /* first do the real part */    
          Naive_SynthesizeX(rdataptr,
                            bw,
                            m,
                            rinvfltres,
                            transpose_seminaive_naive_table[m]);
          
          /* now do the imaginary */    
          Naive_SynthesizeX(idataptr,
                            bw,
                            m,
                            iminvfltres,
                            transpose_seminaive_naive_table[m]);

          /* will store normal, then tranpose before doing inverse fft    */
          memcpy(rfourdata+(m*size), rinvfltres, sizeof(double) * size);
          memcpy(ifourdata+(m*size), iminvfltres, sizeof(double) * size);
        
          /* move to next set of coeffs */
        
          rdataptr += bw-m;
          idataptr += bw-m;
        
        }
    }
  /* closes m loop */
 
  /* now fill in zero values where m = bw (from problem definition) */
  memset(rfourdata + (bw * size), 0, sizeof(double) * size);
  memset(ifourdata + (bw * size), 0, sizeof(double) * size);
  
  /* now if the data is real, we don't have to compute the
     coefficients whose order is less than 0, i.e. since
     the data is real, we know that
     invf-hat(l,-m) = conjugate(invf-hat(l,m)),
     so use that to get the rest of the real data
     
     dataformat =0 -> samples are complex, =1 -> samples real
     
  */

  if(dataformat == 0){
    
    /* now do negative m values */
    
    for (m=bw+1; m<size; m++) 
      {
        /*
          fprintf(stderr,"m = %d\n",-(size-m));
        */
        
        if ( (size-m) < cutoff )
          {
            /* do real part first */
            InvSemiNaiveReduced(rdataptr,
                                bw,
                                size - m,
                                rinvfltres,
                                transpose_seminaive_naive_table[size - m],
                                sin_values,
                                scratchpad,
                                idctPlan);
        
            /* now do imaginary part */
            InvSemiNaiveReduced(idataptr,
                                bw,
                                size - m,
                                iminvfltres,
                                transpose_seminaive_naive_table[size - m],
                                sin_values,
                                scratchpad,
                                idctPlan );

            /* will store normal, then tranpose before doing inverse fft    */
            if ((m % 2) != 0)
              for(i=0; i< size; i++){
                rinvfltres[i] = -rinvfltres[i];
                iminvfltres[i] = -iminvfltres[i];
              }
        
            memcpy(rfourdata + (m*size), rinvfltres, sizeof(double) * size);
            memcpy(ifourdata + (m*size), iminvfltres, sizeof(double) * size);
                
            /* move to next set of coeffs */
            rdataptr += bw-(size-m);
            idataptr += bw-(size-m);
          }
        else
          {
            /* first do the real part */
            Naive_SynthesizeX(rdataptr,
                              bw,
                              size-m,
                              rinvfltres,
                              transpose_seminaive_naive_table[size-m]);
          
            /* now do the imaginary */  
            Naive_SynthesizeX(idataptr,
                              bw,
                              size-m,
                              iminvfltres,
                              transpose_seminaive_naive_table[size-m]);

            /* will store normal, then tranpose before doing inverse fft    */
            if ((m % 2) != 0)
              for(i=0; i< size; i++){
                rinvfltres[i] = -rinvfltres[i];
                iminvfltres[i] = -iminvfltres[i];
              }
        
            memcpy(rfourdata + (m*size), rinvfltres, sizeof(double) * size);
            memcpy(ifourdata + (m*size), iminvfltres, sizeof(double) * size);
        
            /* move to next set of coeffs */
            rdataptr += bw-(size-m);
            idataptr += bw-(size-m);

          }

      } /* closes m loop */
  }
  else {
    for(m = bw + 1; m < size; m++){
      
      memcpy(rfourdata+(m*size), rfourdata+((size-m)*size),
             sizeof(double) * size);
      memcpy(ifourdata+(m*size), ifourdata+((size-m)*size),
             sizeof(double) * size);
      for(i = 0; i < size; i++)
        ifourdata[(m*size)+i] *= -1.0;
    }
  }

  /* normalize */
  tmpA = 1./(sqrt(2.*M_PI) );
  for(i=0;i<4*bw*bw;i++)
    {
      rfourdata[i] *= tmpA ;
      ifourdata[i] *= tmpA ;
    }


  fftw_execute_split_dft( *ifftPlan,
                          ifourdata, rfourdata,
                          idata, rdata );
  /* amscray */
  
}

/************************************************************************/
/*
  Zonal Harmonic transform using seminaive algorithm - used in convolutions

  bw -> bandwidth of problem

  size = 2 * bw

  rdata and idata should be pointers to size x size arrays.
  rres and ires should be pointers to double arrays of size bw.
  
  cos_pml_table contains Legendre coefficients of P(0,l) functions
  and is result of CosPmlTableGen for m = 0;
  FZT_semi only computes spherical harmonics for m=0.

  dataformat =0 -> samples are complex, =1 -> samples real
  
  workspace needed is (12 * bw) 

*/

void FZT_semi_memo(double *rdata, double *idata,
                   double *rres, double *ires,
                   int bw,
                   double *cos_pml_table,
                   double *workspace,
                   int dataformat,
                   fftw_plan *dctPlan,
                   double *weights )
{
  int i, j, size;
  double *r0, *i0, dsize;
  double tmpreal, tmpimag;
  double tmpA ;
  double *scratchpad ;

  size = 2*bw ;

  /* assign memory */
  r0 = workspace;             /* needs (2 * bw) */
  i0 = r0 + (2 * bw);         /* needs (2 * bw) */
  scratchpad = i0 + (2 * bw);   /* needs (4 * bw) */

  /* total workspace = 13*bw */

  dsize = 1.0 / ((double) size);
  tmpA = sqrt( 2.* M_PI );
  dsize *= tmpA ;

  /* compute the m = 0 components */
  for (i=0; i<size; i++) {
    tmpreal = 0.0;
    tmpimag = 0.0;

    for(j=0; j<size; j++) {
      tmpreal += rdata[(i*size)+j];
      tmpimag += idata[(i*size)+j];
    }
    /* normalize */
    r0[i] = tmpreal*dsize;
    i0[i] = tmpimag*dsize;
  }

  /* do the real part */
  SemiNaiveReduced(r0, 
                   bw, 
                   0,
                   rres,
                   scratchpad,
                   cos_pml_table,
                   weights,
                   dctPlan);

  if(dataformat == 0)   /* do imaginary part */
    SemiNaiveReduced(i0, 
                     bw, 
                     0,
                     ires,
                     scratchpad,
                     cos_pml_table,
                     weights,
                     dctPlan);
  else                 /* otherwise set coefficients = 0 */
    memset(ires, 0, sizeof(double) * size);

}

/************************************************************************/
/*
  multiplies harmonic coefficients of a function and a filter.
  See convolution theorem of Driscoll and Healy for details.

  bw -> bandwidth of problem
  size = 2*bw

  datacoeffs should be output of an FST, filtercoeffs the 
  output of an FZT.  There should be (bw * bw) datacoeffs,
  and bw filtercoeffs.
  rres and ires should point to arrays of dimension bw * bw.

*/

void TransMult(double *rdatacoeffs, double *idatacoeffs, 
               double *rfiltercoeffs, double *ifiltercoeffs, 
               double *rres, double *ires,
               int bw)
{

  int m, l, size;
  double *rdptr, *idptr, *rrptr, *irptr;

  size = 2*bw ;

  rdptr = rdatacoeffs;
  idptr = idatacoeffs;
  rrptr = rres;
  irptr = ires;

  for (m=0; m<bw; m++) {
    for (l=m; l<bw; l++) {
      compmult(rfiltercoeffs[l], ifiltercoeffs[l],
               rdptr[l-m], idptr[l-m],
               rrptr[l-m], irptr[l-m]);

      rrptr[l-m] *= sqrt(4*M_PI/(2*l+1));
      irptr[l-m] *= sqrt(4*M_PI/(2*l+1));

    }
    rdptr += bw-m; idptr += bw-m;
    rrptr += bw-m; irptr += bw-m;
  }
  for (m=bw+1; m<size; m++) {
    for (l=size-m; l<bw; l++){
      compmult(rfiltercoeffs[l], ifiltercoeffs[l],
               rdptr[l-size+m], idptr[l-size+m],
               rrptr[l-size+m], irptr[l-size+m]);

      rrptr[l-size+m] *= sqrt(4*M_PI/(2*l+1));
      irptr[l-size+m] *= sqrt(4*M_PI/(2*l+1));

    }
    rdptr += m-bw; idptr += m-bw;
    rrptr += m-bw; irptr += m-bw;
  }

}

/************************************************************************/
/* Here's the big banana
   Convolves two functions defined on the 2-sphere.
   Uses seminaive algorithms for spherical harmonic transforms

   size = 2*bw

   Inputs:

   rdata, idata - (size * size) arrays containing real and
                  imaginary parts of sampled function.
   rfilter, ifilter - (size * size) arrays containing real and
                      imaginary parts of sampled filter function.
   rres, ires - (size * size) arrays containing real and
                  imaginary parts of result function.


   Suggestion - if you want to do multiple convolutions,
   don't keep allocating and freeing space with every call,
   or keep recomputing the spharmonic_pml tables.
   Allocate workspace once before you call this function, then
   just set up pointers as first step of this procedure rather
   than mallocing.  And do the same with the FST, FZT, and InvFST functions.

   ASSUMPTIONS:
   1. data is strictly REAL
   2. will do semi-naive algorithm for ALL orders -> change the cutoff
      value if you want it to be different

   Memory requirements for Conv2Sphere

   Need space for spharmonic tables and local workspace and
   scratchpad space for FST_semi

   Let legendreSize = Reduced_Naive_TableSize(bw,cutoff) +
                      Reduced_SpharmonicTableSize(bw,cutoff)

   Then the workspace needs to be this large:

   2 * legendreSize  +
   8 * (bw*bw)  + 10*bw +
   4 * (bw*bw) + 2*bw

   for a total of

   2 * legendreSize  +
   12 * (bw*bw) + 12*bw ;
   

*/
void Conv2Sphere_semi_memo(double *rdata, double *idata, 
                           double *rfilter, double *ifilter, 
                           double *rres, double *ires, 
                           int bw,
                           double *workspace)
{
  int size, spharmonic_bound ;
  int legendreSize, cutoff ;
  double *frres, *fires, *filtrres, *filtires, *trres, *tires;
  double **spharmonic_pml_table, **transpose_spharmonic_pml_table;
  double *spharmonic_result_space, *transpose_spharmonic_result_space;
  double *scratchpad;

  /* fftw */
  int rank, howmany_rank ;
  fftw_iodim dims[1], howmany_dims[1];

  /* forward transform stuff */
  fftw_plan dctPlan, fftPlan ;
  double *weights ;

  /* inverse transform stuff */
  fftw_plan idctPlan, ifftPlan ;

  size =2*bw ;
  cutoff = bw ;
  legendreSize = Reduced_Naive_TableSize(bw,cutoff) +
    Reduced_SpharmonicTableSize(bw,cutoff) ;

  /* assign space */

  spharmonic_bound = legendreSize ;

  spharmonic_result_space = workspace;          /* needs legendreSize */

  transpose_spharmonic_result_space =
    spharmonic_result_space +  legendreSize ;   /* needs legendreSize */

  frres = transpose_spharmonic_result_space + 
    legendreSize ;                              /* needs (bw*bw) */
  fires = frres + (bw*bw);                      /* needs (bw*bw) */
  trres = fires + (bw*bw);                      /* needs (bw*bw) */
  tires = trres + (bw*bw);                      /* needs (bw*bw) */
  filtrres = tires + (bw*bw);                   /* needs bw */
  filtires = filtrres + bw;                     /* needs bw */
  scratchpad = filtires + bw;                   /* needs (8*bw^2)+(10*bw) */

  /* allocate space, and compute, the weights for this bandwidth */
  weights = (double *) malloc(sizeof(double) * 4 * bw);
  makeweights( bw, weights );

  /* make the fftw plans */

  /* make DCT plans -> note that I will be using the GURU
     interface to execute these plans within the routines*/

  /* forward DCT */
  dctPlan = fftw_plan_r2r_1d( 2*bw, weights, rdata,
                              FFTW_REDFT10, FFTW_ESTIMATE ) ;
      
  /* inverse DCT */
  idctPlan = fftw_plan_r2r_1d( 2*bw, weights, rdata,
                               FFTW_REDFT01, FFTW_ESTIMATE );
  
  /*
    fft "preamble" ;
    note that this plan places the output in a transposed array
  */
  rank = 1 ;
  dims[0].n = 2*bw ;
  dims[0].is = 1 ;
  dims[0].os = 2*bw ;
  howmany_rank = 1 ;
  howmany_dims[0].n = 2*bw ;
  howmany_dims[0].is = 2*bw ;
  howmany_dims[0].os = 1 ;
 
  /* forward fft */
  fftPlan = fftw_plan_guru_split_dft( rank, dims,
                                      howmany_rank, howmany_dims,
                                      rdata, idata,
                                      workspace, workspace+(4*bw*bw),
                                      FFTW_ESTIMATE );

  /*
    now plan for inverse fft - note that this plans assumes
    that I'm working with a transposed array, e.g. the inputs
    for a length 2*bw transform are placed every 2*bw apart,
    the output will be consecutive entries in the array
  */
  rank = 1 ;
  dims[0].n = 2*bw ;
  dims[0].is = 2*bw ;
  dims[0].os = 1 ;
  howmany_rank = 1 ;
  howmany_dims[0].n = 2*bw ;
  howmany_dims[0].is = 1 ;
  howmany_dims[0].os = 2*bw ;

  /* inverse fft */
  ifftPlan = fftw_plan_guru_split_dft( rank, dims,
                                       howmany_rank, howmany_dims,
                                       rdata, idata,
                                       workspace, workspace+(4*bw*bw),
                                       FFTW_ESTIMATE );


  /* precompute the associated Legendre fcts */
  spharmonic_pml_table = 
    Spharmonic_Pml_Table(bw,
                         spharmonic_result_space,
                         scratchpad);

  transpose_spharmonic_pml_table = 
    Transpose_Spharmonic_Pml_Table(spharmonic_pml_table, 
                                   bw,
                                   transpose_spharmonic_result_space,
                                   scratchpad);
  FST_semi_memo(rdata, idata, 
                frres, fires, 
                bw, 
                spharmonic_pml_table,
                scratchpad,
                1,
                bw,
                &dctPlan,
                &fftPlan,
                weights );

  FZT_semi_memo(rfilter, ifilter, 
                filtrres, filtires, 
                bw, 
                spharmonic_pml_table[0],
                scratchpad,
                1,
                &dctPlan,
                weights );

  TransMult(frres, fires, filtrres, filtires, trres, tires, bw);

  InvFST_semi_memo(trres, tires, 
                   rres, ires, 
                   bw,
                   transpose_spharmonic_pml_table,
                   scratchpad,
                   1,
                   bw,
                   &idctPlan,
                   &ifftPlan );

  free( weights ) ;

  /***
      have to free the memory that was allocated in
      Spharmonic_Pml_Table() and
      Transpose_Spharmonic_Pml_Table()
  ***/

  free(spharmonic_pml_table);
  free(transpose_spharmonic_pml_table);

  /* destroy plans */
  fftw_destroy_plan( ifftPlan ) ;
  fftw_destroy_plan( fftPlan ) ;
  fftw_destroy_plan( idctPlan ) ;
  fftw_destroy_plan( dctPlan ) ;
}
Revision:	1287
Committed:	Wed Jun 23 20:18:48 2004 UTC (21 years, 4 months ago) by chrisfen
Content type:	text/plain
File size:	27350 byte(s)
Log Message:	Major progress towards inclusion of spherical harmonic transform capability - still having some build issues...