| 1 | #include "machdefs.h"
|
|---|
| 2 | #include <iostream>
|
|---|
| 3 | #include <math.h>
|
|---|
| 4 | #include <complex>
|
|---|
| 5 | #include "sphericaltransformserver.h"
|
|---|
| 6 | #include "tvector.h"
|
|---|
| 7 | #include "nbrandom.h"
|
|---|
| 8 | #include "nbmath.h"
|
|---|
| 9 | #include "timing.h"
|
|---|
| 10 | //#include "spherehealpix.h"
|
|---|
| 11 |
|
|---|
| 12 |
|
|---|
| 13 | /*! \class SOPHYA::SphericalTransformServer
|
|---|
| 14 |
|
|---|
| 15 | Class for performing analysis and synthesis of sky maps using spin-0 or spin-2 spherical harmonics.
|
|---|
| 16 |
|
|---|
| 17 | Maps must be SOPHYA SphericalMaps (SphereGorski or SphereThetaPhi).
|
|---|
| 18 |
|
|---|
| 19 | Temperature and polarization (Stokes parameters) can be developped on spherical harmonics :
|
|---|
| 20 | \f[
|
|---|
| 21 | \frac{\Delta T}{T}(\hat{n})=\sum_{lm}a_{lm}^TY_l^m(\hat{n})
|
|---|
| 22 | \f]
|
|---|
| 23 | \f[
|
|---|
| 24 | Q(\hat{n})=\frac{1}{\sqrt{2}}\sum_{lm}N_l\left(a_{lm}^EW_{lm}(\hat{n})+a_{lm}^BX_{lm}(\hat{n})\right)
|
|---|
| 25 | \f]
|
|---|
| 26 | \f[
|
|---|
| 27 | U(\hat{n})=-\frac{1}{\sqrt{2}}\sum_{lm}N_l\left(a_{lm}^EX_{lm}(\hat{n})-a_{lm}^BW_{lm}(\hat{n})\right)
|
|---|
| 28 | \f]
|
|---|
| 29 | \f[
|
|---|
| 30 | \left(Q \pm iU\right)(\hat{n})=\sum_{lm}a_{\pm 2lm}\, _{\pm 2}Y_l^m(\hat{n})
|
|---|
| 31 | \f]
|
|---|
| 32 |
|
|---|
| 33 | \f[
|
|---|
| 34 | Y_l^m(\hat{n})=\lambda_l^m(\theta)e^{im\phi}
|
|---|
| 35 | \f]
|
|---|
| 36 | \f[
|
|---|
| 37 | _{\pm}Y_l^m(\hat{n})=_{\pm}\lambda_l^m(\theta)e^{im\phi}
|
|---|
| 38 | \f]
|
|---|
| 39 | \f[
|
|---|
| 40 | W_{lm}(\hat{n})=\frac{1}{N_l}\,_{w}\lambda_l^m(\theta)e^{im\phi}
|
|---|
| 41 | \f]
|
|---|
| 42 | \f[
|
|---|
| 43 | X_{lm}(\hat{n})=\frac{-i}{N_l}\,_{x}\lambda_l^m(\theta)e^{im\phi}
|
|---|
| 44 | \f]
|
|---|
| 45 |
|
|---|
| 46 | (see LambdaLMBuilder, LambdaPMBuilder, LambdaWXBuilder classes)
|
|---|
| 47 |
|
|---|
| 48 | power spectra :
|
|---|
| 49 |
|
|---|
| 50 | \f[
|
|---|
| 51 | C_l^T=\frac{1}{2l+1}\sum_{m=0}^{+ \infty }\left|a_{lm}^T\right|^2=\langle\left|a_{lm}^T\right|^2\rangle
|
|---|
| 52 | \f]
|
|---|
| 53 | \f[
|
|---|
| 54 | C_l^E=\frac{1}{2l+1}\sum_{m=0}^{+\infty}\left|a_{lm}^E\right|^2=\langle\left|a_{lm}^E\right|^2\rangle
|
|---|
| 55 | \f]
|
|---|
| 56 | \f[
|
|---|
| 57 | C_l^B=\frac{1}{2l+1}\sum_{m=0}^{+\infty}\left|a_{lm}^B\right|^2=\langle\left|a_{lm}^B\right|^2\rangle
|
|---|
| 58 | \f]
|
|---|
| 59 |
|
|---|
| 60 | \arg
|
|---|
| 61 | \b Synthesis : Get temperature and polarization maps from \f$a_{lm}\f$ coefficients or from power spectra, (methods GenerateFrom...).
|
|---|
| 62 |
|
|---|
| 63 | \b Temperature:
|
|---|
| 64 | \f[
|
|---|
| 65 | \frac{\Delta T}{T}(\hat{n})=\sum_{lm}a_{lm}^TY_l^m(\hat{n}) = \sum_{-\infty}^{+\infty}b_m(\theta)e^{im\phi}
|
|---|
| 66 | \f]
|
|---|
| 67 |
|
|---|
| 68 | with
|
|---|
| 69 | \f[
|
|---|
| 70 | b_m(\theta)=\sum_{l=\left|m\right|}^{+\infty}a_{lm}^T\lambda_l^m(\theta)
|
|---|
| 71 | \f]
|
|---|
| 72 |
|
|---|
| 73 | \b Polarisation
|
|---|
| 74 | \f[
|
|---|
| 75 | Q \pm iU = \sum_{-\infty}^{+\infty}b_m^{\pm}(\theta)e^{im\phi}
|
|---|
| 76 | \f]
|
|---|
| 77 |
|
|---|
| 78 | where :
|
|---|
| 79 | \f[
|
|---|
| 80 | b_m^{\pm}(\theta) = \sum_{l=\left|m\right|}^{+\infty}a_{\pm 2lm}\,_{\pm}\lambda_l^m(\theta)
|
|---|
| 81 | \f]
|
|---|
| 82 |
|
|---|
| 83 | or :
|
|---|
| 84 | \f[
|
|---|
| 85 | Q = \sum_{-\infty}^{+\infty}b_m^{Q}(\theta)e^{im\phi}
|
|---|
| 86 | \f]
|
|---|
| 87 | \f[
|
|---|
| 88 | U = \sum_{-\infty}^{+\infty}b_m^{U}(\theta)e^{im\phi}
|
|---|
| 89 | \f]
|
|---|
| 90 |
|
|---|
| 91 | where:
|
|---|
| 92 | \f[
|
|---|
| 93 | b_m^{Q}(\theta) = \frac{1}{\sqrt{2}}\sum_{l=\left|m\right|}^{+\infty}\left(a_{lm}^E\,_{w}\lambda_l^m(\theta)-ia_{lm}^B\,_{x}\lambda_l^m(\theta)\right)
|
|---|
| 94 | \f]
|
|---|
| 95 | \f[
|
|---|
| 96 | b_m^{U}(\theta) = \frac{1}{\sqrt{2}}\sum_{l=\left|m\right|}^{+\infty}\left(ia_{lm}^E\,_{x}\lambda_l^m(\theta)+a_{lm}^B\,_{w}\lambda_l^m(\theta)\right)
|
|---|
| 97 | \f]
|
|---|
| 98 |
|
|---|
| 99 | Since the pixelization provides "slices" with constant \f$\theta\f$ and \f$\phi\f$ equally distributed on \f$2\pi\f$ \f$\frac{\Delta T}{T}\f$, \f$Q\f$,\f$U\f$ can be computed by FFT.
|
|---|
| 100 |
|
|---|
| 101 |
|
|---|
| 102 | \arg
|
|---|
| 103 | \b Analysis : Get \f$a_{lm}\f$ coefficients or power spectra from temperature and polarization maps (methods DecomposeTo...).
|
|---|
| 104 |
|
|---|
| 105 | \b Temperature:
|
|---|
| 106 | \f[
|
|---|
| 107 | a_{lm}^T=\int\frac{\Delta T}{T}(\hat{n})Y_l^{m*}(\hat{n})d\hat{n}
|
|---|
| 108 | \f]
|
|---|
| 109 |
|
|---|
| 110 | approximated as :
|
|---|
| 111 | \f[
|
|---|
| 112 | a_{lm}^T=\sum_{\theta_k}\omega_kC_m(\theta_k)\lambda_l^m(\theta_k)
|
|---|
| 113 | \f]
|
|---|
| 114 | where :
|
|---|
| 115 | \f[
|
|---|
| 116 | C_m (\theta _k)=\sum_{\phi _{k\prime}}\frac{\Delta T}{T}(\theta _k,\phi_{k\prime})e^{-im\phi _{k\prime}}
|
|---|
| 117 | \f]
|
|---|
| 118 | Since the pixelization provides "slices" with constant \f$\theta\f$ and \f$\phi\f$ equally distributed on \f$2\pi\f$ (\f$\omega_k\f$ is the solid angle of each pixel of the slice \f$\theta_k\f$) \f$C_m\f$ can be computed by FFT.
|
|---|
| 119 |
|
|---|
| 120 | \b polarisation:
|
|---|
| 121 |
|
|---|
| 122 | \f[
|
|---|
| 123 | a_{\pm 2lm}=\sum_{\theta_k}\omega_kC_m^{\pm}(\theta_k)\,_{\pm}\lambda_l^m(\theta_k)
|
|---|
| 124 | \f]
|
|---|
| 125 | where :
|
|---|
| 126 | \f[
|
|---|
| 127 | C_m^{\pm} (\theta _k)=\sum_{\phi _{k\prime}}\left(Q \pm iU\right)(\theta _k,\phi_{k\prime})e^{-im\phi _{k\prime}}
|
|---|
| 128 | \f]
|
|---|
| 129 | or :
|
|---|
| 130 |
|
|---|
| 131 | \f[
|
|---|
| 132 | a_{lm}^E=\frac{1}{\sqrt{2}}\sum_{\theta_k}\omega_k\left(C_m^{Q}(\theta_k)\,_{w}\lambda_l^m(\theta_k)-iC_m^{U}(\theta_k)\,_{x}\lambda_l^m(\theta_k)\right)
|
|---|
| 133 | \f]
|
|---|
| 134 | \f[
|
|---|
| 135 | a_{lm}^B=\frac{1}{\sqrt{2}}\sum_{\theta_k}\omega_k\left(iC_m^{Q}(\theta_k)\,_{x}\lambda_l^m(\theta_k)+C_m^{U}(\theta_k)\,_{w}\lambda_l^m(\theta_k)\right)
|
|---|
| 136 | \f]
|
|---|
| 137 |
|
|---|
| 138 | where :
|
|---|
| 139 | \f[
|
|---|
| 140 | C_m^{Q} (\theta _k)=\sum_{\phi _{k\prime}}Q(\theta _k,\phi_{k\prime})e^{-im\phi _{k\prime}}
|
|---|
| 141 | \f]
|
|---|
| 142 | \f[
|
|---|
| 143 | C_m^{U} (\theta _k)=\sum_{\phi _{k\prime}}U(\theta _k,\phi_{k\prime})e^{-im\phi _{k\prime}}
|
|---|
| 144 | \f]
|
|---|
| 145 |
|
|---|
| 146 | */
|
|---|
| 147 |
|
|---|
| 148 | /*! \fn void SOPHYA::SphericalTransformServer::GenerateFromAlm( SphericalMap<T>& map, int_4 pixelSizeIndex, const Alm<T>& alm) const
|
|---|
| 149 |
|
|---|
| 150 | synthesis of a temperature map from Alm coefficients
|
|---|
| 151 | */
|
|---|
| 152 | template<class T>
|
|---|
| 153 | void SphericalTransformServer<T>::GenerateFromAlm( SphericalMap<T>& map, int_4 pixelSizeIndex, const Alm<T>& alm) const
|
|---|
| 154 | {
|
|---|
| 155 | /*=======================================================================
|
|---|
| 156 | computes a map from its alm for the HEALPIX pixelisation
|
|---|
| 157 | map(theta,phi) = sum_l_m a_lm Y_lm(theta,phi)
|
|---|
| 158 | = sum_m {e^(i*m*phi) sum_l a_lm*lambda_lm(theta)}
|
|---|
| 159 |
|
|---|
| 160 | where Y_lm(theta,phi) = lambda(theta) * e^(i*m*phi)
|
|---|
| 161 |
|
|---|
| 162 | * the recurrence of Ylm is the standard one (cf Num Rec)
|
|---|
| 163 | * the sum over m is done by FFT
|
|---|
| 164 |
|
|---|
| 165 | =======================================================================*/
|
|---|
| 166 | int_4 nlmax=alm.Lmax();
|
|---|
| 167 | int_4 nmmax=nlmax;
|
|---|
| 168 | int_4 nsmax=0;
|
|---|
| 169 | // le Resize est suppose mettre a zero
|
|---|
| 170 | map.Resize(pixelSizeIndex);
|
|---|
| 171 | string sphere_type=map.TypeOfMap();
|
|---|
| 172 | int premiereTranche = 0;
|
|---|
| 173 | int derniereTranche = map.NbThetaSlices()-1;
|
|---|
| 174 | if (sphere_type.substr(0,4) == "RING")
|
|---|
| 175 | {
|
|---|
| 176 | nsmax=map.SizeIndex();
|
|---|
| 177 | }
|
|---|
| 178 | else
|
|---|
| 179 | {
|
|---|
| 180 | // pour une sphere Gorski le nombre de pixels est 12*nsmax**2
|
|---|
| 181 | // on calcule une quantite equivalente a nsmax pour la sphere-theta-phi
|
|---|
| 182 | // en vue de l'application du critere Healpix : nlmax<=3*nsmax-1
|
|---|
| 183 | // c'est approximatif ; a raffiner.
|
|---|
| 184 | if (sphere_type.substr(0,6) == "TETAFI")
|
|---|
| 185 | {
|
|---|
| 186 | nsmax=(int_4)sqrt(map.NbPixels()/12.);
|
|---|
| 187 | premiereTranche++;
|
|---|
| 188 | derniereTranche--;
|
|---|
| 189 | }
|
|---|
| 190 | else
|
|---|
| 191 | {
|
|---|
| 192 | cout << " unknown type of sphere : " << sphere_type << endl;
|
|---|
| 193 | throw IOExc(" unknown type of sphere: " + (string)sphere_type );
|
|---|
| 194 | }
|
|---|
| 195 | // cout << "GenerateFromAlm: the sphere is of type : " << sphere_type << endl;
|
|---|
| 196 | // cout << "GenerateFromAlm: size index (nside) of the sphere= " << nsmax << endl;
|
|---|
| 197 | // cout << "GenerateFromAlm: nlmax (from Alm) = " << nlmax << endl;
|
|---|
| 198 | // if (nlmax>3*nsmax-1)
|
|---|
| 199 | // {
|
|---|
| 200 | // cout << "GenerateFromAlm: nlmax should be <= 3*nside-1" << endl;
|
|---|
| 201 | // if (strncmp(sphere_type,"TETAFI",6) == 0)
|
|---|
| 202 | // {
|
|---|
| 203 | // cout << "GenerateFromAlm: nlmax should be <= 3*nside-1" << endl;
|
|---|
| 204 | // cout << " (for this criterium, nsmax is computed as sqrt(nbPixels/12))" << endl;
|
|---|
| 205 | // }
|
|---|
| 206 | //}
|
|---|
| 207 | }
|
|---|
| 208 | Bm<complex<T> > b_m_theta(nmmax);
|
|---|
| 209 |
|
|---|
| 210 | // map.Resize(nsmax);
|
|---|
| 211 |
|
|---|
| 212 |
|
|---|
| 213 | // pour chaque tranche en theta
|
|---|
| 214 | for (int_4 ith = premiereTranche; ith <= derniereTranche;ith++)
|
|---|
| 215 | {
|
|---|
| 216 | int_4 nph;
|
|---|
| 217 | r_8 phi0;
|
|---|
| 218 | r_8 theta;
|
|---|
| 219 | TVector<int_4> pixNumber;
|
|---|
| 220 | TVector<T> datan;
|
|---|
| 221 |
|
|---|
| 222 | map.GetThetaSlice(ith,theta,phi0, pixNumber,datan);
|
|---|
| 223 | nph = pixNumber.NElts();
|
|---|
| 224 |
|
|---|
| 225 | // -----------------------------------------------------
|
|---|
| 226 | // for each theta, and each m, computes
|
|---|
| 227 | // b(m,theta) = sum_over_l>m (lambda_l_m(theta) * a_l_m)
|
|---|
| 228 | // ------------------------------------------------------
|
|---|
| 229 | LambdaLMBuilder lb(theta,nlmax,nmmax);
|
|---|
| 230 | // somme sur m de 0 a l'infini
|
|---|
| 231 | int m;
|
|---|
| 232 | for (m = 0; m <= nmmax; m++)
|
|---|
| 233 | {
|
|---|
| 234 | b_m_theta(m) = (T)( lb.lamlm(m,m) ) * alm(m,m);
|
|---|
| 235 | for (int l = m+1; l<= nlmax; l++)
|
|---|
| 236 | {
|
|---|
| 237 | b_m_theta(m) += (T)( lb.lamlm(l,m) ) * alm(l,m);
|
|---|
| 238 | }
|
|---|
| 239 | }
|
|---|
| 240 | // obtains the negative m of b(m,theta) (= complex conjugate)
|
|---|
| 241 |
|
|---|
| 242 | for (m=1;m<=nmmax;m++)
|
|---|
| 243 | {
|
|---|
| 244 | b_m_theta(-m) = conj(b_m_theta(m));
|
|---|
| 245 | }
|
|---|
| 246 | // ---------------------------------------------------------------
|
|---|
| 247 | // sum_m b(m,theta)*exp(i*m*phi) -> f(phi,theta)
|
|---|
| 248 | // ---------------------------------------------------------------*/
|
|---|
| 249 |
|
|---|
| 250 |
|
|---|
| 251 | if (sphere_type.substr(0,4) == "RING")
|
|---|
| 252 | {
|
|---|
| 253 | TVector<T> Temp = RfourierSynthesisFromB(b_m_theta,nph,phi0);
|
|---|
| 254 | for (int i=0;i< nph;i++) map(pixNumber(i))=Temp(i);
|
|---|
| 255 | }
|
|---|
| 256 | else
|
|---|
| 257 | // pour des pixelisations quelconques (autres que HEALPix
|
|---|
| 258 | // nph n'est pas toujours pair
|
|---|
| 259 | // ca fait des problemes pour les transformees de Fourier
|
|---|
| 260 | // car le server de TF ajuste la longueur du vecteur reel
|
|---|
| 261 | // en sortie de TF, bref, la securite veut qu'on prenne une
|
|---|
| 262 | // TF complexe
|
|---|
| 263 | {
|
|---|
| 264 | TVector<complex<T> > Temp = fourierSynthesisFromB(b_m_theta,nph,phi0);
|
|---|
| 265 | for (int i=0;i< nph;i++) map(pixNumber(i))=Temp(i).real();
|
|---|
| 266 | }
|
|---|
| 267 | }
|
|---|
| 268 | }
|
|---|
| 269 |
|
|---|
| 270 |
|
|---|
| 271 |
|
|---|
| 272 | /*! \fn TVector< complex<T> > SOPHYA::SphericalTransformServer::fourierSynthesisFromB(const Bm<complex<T> >& b_m, int_4 nph, r_8 phi0) const
|
|---|
| 273 |
|
|---|
| 274 | \return a vector with nph elements which are sums :\f$\sum_{m=-mmax}^{mmax}b_m(\theta)e^{im\varphi}\f$ for nph values of \f$\varphi\f$ regularly distributed in \f$[0,\pi]\f$ ( calculated by FFT)
|
|---|
| 275 |
|
|---|
| 276 | The object b_m (\f$b_m\f$) of the class Bm is a special vector which index goes from -mmax to mmax.
|
|---|
| 277 | */
|
|---|
| 278 | template<class T>
|
|---|
| 279 | TVector< complex<T> > SphericalTransformServer<T>::fourierSynthesisFromB(const Bm<complex<T> >& b_m, int_4 nph, r_8 phi0) const
|
|---|
| 280 | {
|
|---|
| 281 | /*=======================================================================
|
|---|
| 282 | dataout(j) = sum_m datain(m) * exp(i*m*phi(j))
|
|---|
| 283 | with phi(j) = j*2pi/nph + kphi0*pi/nph and kphi0 =0 or 1
|
|---|
| 284 |
|
|---|
| 285 | as the set of frequencies {m} is larger than nph,
|
|---|
| 286 | we wrap frequencies within {0..nph-1}
|
|---|
| 287 | ie m = k*nph + m' with m' in {0..nph-1}
|
|---|
| 288 | then
|
|---|
| 289 | noting bw(m') = exp(i*m'*phi0)
|
|---|
| 290 | * sum_k (datain(k*nph+m') exp(i*k*pi*kphi0))
|
|---|
| 291 | with bw(nph-m') = CONJ(bw(m')) (if datain(-m) = CONJ(datain(m)))
|
|---|
| 292 | dataout(j) = sum_m' [ bw(m') exp (i*j*m'*2pi/nph) ]
|
|---|
| 293 | = Fourier Transform of bw
|
|---|
| 294 | is real
|
|---|
| 295 |
|
|---|
| 296 | NB nph is not necessarily a power of 2
|
|---|
| 297 |
|
|---|
| 298 | =======================================================================*/
|
|---|
| 299 | //**********************************************************************
|
|---|
| 300 | // pour une valeur de phi (indexee par j) la temperature est la transformee
|
|---|
| 301 | // de Fourier de bm (somme sur m de -nmax a +nmmax de bm*exp(i*m*phi)).
|
|---|
| 302 | // on demande nph (nombre de pixels sur la tranche) valeurs de transformees, pour nph valeurs de phi, regulierement reparties sur 2*pi. On a:
|
|---|
| 303 | // DT/T(j) = sum_m b(m) * exp(i*m*phi(j))
|
|---|
| 304 | // sommation de -infini a +infini, en fait limitee a -nmamx, +nmmax
|
|---|
| 305 | // On pose m=k*nph + m', avec m' compris entre 0 et nph-1. Alors :
|
|---|
| 306 | // DT/T(j) = somme_k somme_m' b(k*nph + m')*exp(i*(k*nph + m')*phi(j))
|
|---|
| 307 | // somme_k : de -infini a +infini
|
|---|
| 308 | // somme_m' : de 0 a nph-1
|
|---|
| 309 | // On echange les sommations :
|
|---|
| 310 | // DT/T(j) = somme_k (exp(i*m'*phi(j)) somme_m' b(k*nph + m')*exp(i*(k*nph*phi(j))
|
|---|
| 311 | // mais phi(j) est un multiple entier de 2*pi/nph, la seconde exponentielle
|
|---|
| 312 | // vaut 1.
|
|---|
| 313 | // Il reste a calculer les transformees de Fourier de somme_m' b(k*nph + m')
|
|---|
| 314 | // si phi0 n'est pas nul, il y a juste un decalage a faire.
|
|---|
| 315 | //**********************************************************************
|
|---|
| 316 |
|
|---|
| 317 | TVector< complex<T> > bw(nph);
|
|---|
| 318 | TVector< complex<T> > dataout(nph);
|
|---|
| 319 | TVector< complex<T> > data(nph);
|
|---|
| 320 |
|
|---|
| 321 |
|
|---|
| 322 | for (int kk=0; kk<bw.NElts(); kk++) bw(kk)=(T)0.;
|
|---|
| 323 | int m;
|
|---|
| 324 | for (m=-b_m.Mmax();m<=-1;m++)
|
|---|
| 325 | {
|
|---|
| 326 | int maux=m;
|
|---|
| 327 | while (maux<0) maux+=nph;
|
|---|
| 328 | int iw=maux%nph;
|
|---|
| 329 | double aux=(m-iw)*phi0;
|
|---|
| 330 | bw(iw) += b_m(m) * complex<T>( (T)cos(aux),(T)sin(aux) ) ;
|
|---|
| 331 | }
|
|---|
| 332 | for (m=0;m<=b_m.Mmax();m++)
|
|---|
| 333 | {
|
|---|
| 334 | // int iw=((m % nph) +nph) % nph; //between 0 and nph = m'
|
|---|
| 335 | int iw=m%nph;
|
|---|
| 336 | double aux=(m-iw)*phi0;
|
|---|
| 337 | bw(iw)+=b_m(m) * complex<T>( (T)cos(aux),(T)sin(aux) );
|
|---|
| 338 | }
|
|---|
| 339 |
|
|---|
| 340 | // applies the shift in position <-> phase factor in Fourier space
|
|---|
| 341 | for (int mprime=0; mprime < nph; mprime++)
|
|---|
| 342 | {
|
|---|
| 343 | complex<double> aux(cos(mprime*phi0),sin(mprime*phi0));
|
|---|
| 344 | data(mprime)=bw(mprime)*
|
|---|
| 345 | (complex<T>)(complex<double>(cos(mprime*phi0),sin(mprime*phi0)));
|
|---|
| 346 | }
|
|---|
| 347 |
|
|---|
| 348 | //sortie.ReSize(nph);
|
|---|
| 349 | TVector< complex<T> > sortie(nph);
|
|---|
| 350 |
|
|---|
| 351 | fftIntfPtr_-> FFTBackward(data, sortie);
|
|---|
| 352 |
|
|---|
| 353 | return sortie;
|
|---|
| 354 | }
|
|---|
| 355 |
|
|---|
| 356 | //********************************************
|
|---|
| 357 | /*! \fn TVector<T> SOPHYA::SphericalTransformServer::RfourierSynthesisFromB(const Bm<complex<T> >& b_m, int_4 nph, r_8 phi0) const
|
|---|
| 358 |
|
|---|
| 359 | same as fourierSynthesisFromB, but return a real vector, taking into account the fact that b(-m) is conjugate of b(m) */
|
|---|
| 360 | template<class T>
|
|---|
| 361 | TVector<T> SphericalTransformServer<T>::RfourierSynthesisFromB(const Bm<complex<T> >& b_m, int_4 nph, r_8 phi0) const
|
|---|
| 362 | {
|
|---|
| 363 | /*=======================================================================
|
|---|
| 364 | dataout(j) = sum_m datain(m) * exp(i*m*phi(j))
|
|---|
| 365 | with phi(j) = j*2pi/nph + kphi0*pi/nph and kphi0 =0 or 1
|
|---|
| 366 |
|
|---|
| 367 | as the set of frequencies {m} is larger than nph,
|
|---|
| 368 | we wrap frequencies within {0..nph-1}
|
|---|
| 369 | ie m = k*nph + m' with m' in {0..nph-1}
|
|---|
| 370 | then
|
|---|
| 371 | noting bw(m') = exp(i*m'*phi0)
|
|---|
| 372 | * sum_k (datain(k*nph+m') exp(i*k*pi*kphi0))
|
|---|
| 373 | with bw(nph-m') = CONJ(bw(m')) (if datain(-m) = CONJ(datain(m)))
|
|---|
| 374 | dataout(j) = sum_m' [ bw(m') exp (i*j*m'*2pi/nph) ]
|
|---|
| 375 | = Fourier Transform of bw
|
|---|
| 376 | is real
|
|---|
| 377 |
|
|---|
| 378 | NB nph is not necessarily a power of 2
|
|---|
| 379 |
|
|---|
| 380 | =======================================================================*/
|
|---|
| 381 | //**********************************************************************
|
|---|
| 382 | // pour une valeur de phi (indexee par j) la temperature est la transformee
|
|---|
| 383 | // de Fourier de bm (somme sur m de -nmax a +nmmax de bm*exp(i*m*phi)).
|
|---|
| 384 | // on demande nph (nombre de pixels sur la tranche) valeurs de transformees, pour nph valeurs de phi, regulierement reparties sur 2*pi. On a:
|
|---|
| 385 | // DT/T(j) = sum_m b(m) * exp(i*m*phi(j))
|
|---|
| 386 | // sommation de -infini a +infini, en fait limitee a -nmamx, +nmmax
|
|---|
| 387 | // On pose m=k*nph + m', avec m' compris entre 0 et nph-1. Alors :
|
|---|
| 388 | // DT/T(j) = somme_k somme_m' b(k*nph + m')*exp(i*(k*nph + m')*phi(j))
|
|---|
| 389 | // somme_k : de -infini a +infini
|
|---|
| 390 | // somme_m' : de 0 a nph-1
|
|---|
| 391 | // On echange les sommations :
|
|---|
| 392 | // DT/T(j) = somme_m' (exp(i*m'*phi(j)) somme_k b(k*nph + m')*exp(i*(k*nph*phi(j))
|
|---|
| 393 | // mais phi(j) est un multiple entier de 2*pi/nph, la seconde exponentielle
|
|---|
| 394 | // vaut 1.
|
|---|
| 395 | // Il reste a calculer les transformees de Fourier de somme_k b(k*nph + m')
|
|---|
| 396 | // si phi0 n'est pas nul, il y a juste un decalage a faire.
|
|---|
| 397 | //**********************************************************************
|
|---|
| 398 | TVector< complex<T> > bw(nph);
|
|---|
| 399 | TVector< complex<T> > dataout(nph);
|
|---|
| 400 | TVector< complex<T> > data(nph/2+1);
|
|---|
| 401 |
|
|---|
| 402 |
|
|---|
| 403 | for (int kk=0; kk<bw.NElts(); kk++) bw(kk)=(T)0.;
|
|---|
| 404 | int m;
|
|---|
| 405 | for (m=-b_m.Mmax();m<=-1;m++)
|
|---|
| 406 | {
|
|---|
| 407 | int maux=m;
|
|---|
| 408 | while (maux<0) maux+=nph;
|
|---|
| 409 | int iw=maux%nph;
|
|---|
| 410 | double aux=(m-iw)*phi0;
|
|---|
| 411 | bw(iw) += b_m(m) * complex<T>( (T)cos(aux),(T)sin(aux) ) ;
|
|---|
| 412 | }
|
|---|
| 413 | for (m=0;m<=b_m.Mmax();m++)
|
|---|
| 414 | {
|
|---|
| 415 | // int iw=((m % nph) +nph) % nph; //between 0 and nph = m'
|
|---|
| 416 | int iw=m%nph;
|
|---|
| 417 | double aux=(m-iw)*phi0;
|
|---|
| 418 | bw(iw)+=b_m(m) * complex<T>( (T)cos(aux),(T)sin(aux) );
|
|---|
| 419 | }
|
|---|
| 420 |
|
|---|
| 421 | // applies the shift in position <-> phase factor in Fourier space
|
|---|
| 422 | // cout << " TF : nph= " << nph << " vec. entree " << data.Size() << endl;
|
|---|
| 423 | for (int mprime=0; mprime <= nph/2; mprime++)
|
|---|
| 424 | {
|
|---|
| 425 | complex<double> aux(cos(mprime*phi0),sin(mprime*phi0));
|
|---|
| 426 | data(mprime)=bw(mprime)*
|
|---|
| 427 | (complex<T>)(complex<double>(cos(mprime*phi0),sin(mprime*phi0)));
|
|---|
| 428 | }
|
|---|
| 429 |
|
|---|
| 430 | TVector<T> sortie;
|
|---|
| 431 | fftIntfPtr_-> FFTBackward(data, sortie);
|
|---|
| 432 |
|
|---|
| 433 | return sortie;
|
|---|
| 434 | }
|
|---|
| 435 | //*******************************************
|
|---|
| 436 |
|
|---|
| 437 | /*! \fn Alm<T> SOPHYA::SphericalTransformServer::DecomposeToAlm(const SphericalMap<T>& map, int_4 nlmax, r_8 cos_theta_cut) const
|
|---|
| 438 |
|
|---|
| 439 | \return the Alm coefficients from analysis of a temperature map.
|
|---|
| 440 |
|
|---|
| 441 | \param<nlmax> : maximum value of the l index
|
|---|
| 442 |
|
|---|
| 443 | \param<cos_theta_cut> : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 444 |
|
|---|
| 445 | */
|
|---|
| 446 | template<class T>
|
|---|
| 447 | void SphericalTransformServer<T>::DecomposeToAlm(const SphericalMap<T>& map, Alm<T>& alm, int_4 nlmax, r_8 cos_theta_cut) const
|
|---|
| 448 | {
|
|---|
| 449 | DecomposeToAlm(const_cast< SphericalMap<T>& >(map), alm, nlmax, cos_theta_cut, 0);
|
|---|
| 450 | }
|
|---|
| 451 | //*******************************************
|
|---|
| 452 |
|
|---|
| 453 | /*! \fn Alm<T> SOPHYA::SphericalTransformServer::DecomposeToAlm(const SphericalMap<T>& map, int_4 nlmax, r_8 cos_theta_cut, int iterationOrder) const
|
|---|
| 454 |
|
|---|
| 455 | \return the Alm coefficients from analysis of a temperature map. THE MAP CAN BE MODIFIED (if iterationOrder >0)
|
|---|
| 456 |
|
|---|
| 457 | \param<nlmax> : maximum value of the l index
|
|---|
| 458 |
|
|---|
| 459 | \param<cos_theta_cut> : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 460 |
|
|---|
| 461 | \param<iterationOrder> : 1,2,3,4.... order of an iterative analysis. (Default : 0 -> standard analysis). If iterationOrder is not null, the method works with SphereHEALPix but NOT WITH SphereThetaPhi maps ! */
|
|---|
| 462 | template<class T>
|
|---|
| 463 | void SphericalTransformServer<T>::DecomposeToAlm(SphericalMap<T>& map, Alm<T>& alm, int_4 nlmax, r_8 cos_theta_cut, int iterationOrder) const
|
|---|
| 464 | {
|
|---|
| 465 | int_4 nmmax = nlmax;
|
|---|
| 466 | // PrtTim("appel carteVersAlm");
|
|---|
| 467 | carteVersAlm(map, nlmax, cos_theta_cut, alm);
|
|---|
| 468 | // PrtTim("retour carteVersAlm");
|
|---|
| 469 | if (iterationOrder > 0)
|
|---|
| 470 | {
|
|---|
| 471 | TVector<int_4> fact(iterationOrder+2);
|
|---|
| 472 | fact(0) = 1;
|
|---|
| 473 | int k;
|
|---|
| 474 | for (k=1; k <= iterationOrder+1; k++)
|
|---|
| 475 | {
|
|---|
| 476 | fact(k) = fact(k-1)*k;
|
|---|
| 477 | }
|
|---|
| 478 | Alm<T> alm2(alm);
|
|---|
| 479 | T Tzero = (T)0.;
|
|---|
| 480 | complex<T> complexZero = complex<T>(Tzero, Tzero);
|
|---|
| 481 | alm = complexZero;
|
|---|
| 482 | int signe = 1;
|
|---|
| 483 | int nbIteration = iterationOrder+1;
|
|---|
| 484 | for (k=1; k <= nbIteration; k++)
|
|---|
| 485 | {
|
|---|
| 486 | T facMult = (T)(0.5*signe*fact(iterationOrder)*(2*nbIteration-k)/(fact(k)*fact(nbIteration-k)));
|
|---|
| 487 | for (int m = 0; m <= nmmax; m++)
|
|---|
| 488 | {
|
|---|
| 489 | for (int l = m; l<= nlmax; l++)
|
|---|
| 490 | {
|
|---|
| 491 | alm(l,m) += facMult*alm2(l,m);
|
|---|
| 492 | }
|
|---|
| 493 | }
|
|---|
| 494 | if (k == nbIteration) break;
|
|---|
| 495 | signe = -signe;
|
|---|
| 496 | for (int k=0; k< map.NbPixels(); k++) map(k) = (T)0.;
|
|---|
| 497 | // synthetize a map from the estimated alm
|
|---|
| 498 | // PrtTim("appel GenerateFromAlm");
|
|---|
| 499 | GenerateFromAlm( map, map.SizeIndex(), alm2);
|
|---|
| 500 | // PrtTim("retour GenerateFromAlm");
|
|---|
| 501 | alm2 = complexZero;
|
|---|
| 502 | // analyse the new map
|
|---|
| 503 | // PrtTim("appel carteVersAlm");
|
|---|
| 504 | carteVersAlm(map, nlmax, cos_theta_cut, alm2);
|
|---|
| 505 | // PrtTim("retour carteVersAlm");
|
|---|
| 506 | }
|
|---|
| 507 | }
|
|---|
| 508 | }
|
|---|
| 509 |
|
|---|
| 510 | template<class T>
|
|---|
| 511 | void SphericalTransformServer<T>::carteVersAlm(const SphericalMap<T>& map, int_4 nlmax, r_8 cos_theta_cut, Alm<T>& alm) const
|
|---|
| 512 | {
|
|---|
| 513 |
|
|---|
| 514 | /*-----------------------------------------------------------------------
|
|---|
| 515 | computes the integral in phi : phas_m(theta)
|
|---|
| 516 | for each parallele from north to south pole
|
|---|
| 517 | -----------------------------------------------------------------------*/
|
|---|
| 518 | TVector<T> data;
|
|---|
| 519 | TVector<int_4> pixNumber;
|
|---|
| 520 | int_4 nmmax = nlmax;
|
|---|
| 521 | TVector< complex<T> > phase(nmmax+1);
|
|---|
| 522 |
|
|---|
| 523 | alm.ReSizeToLmax(nlmax);
|
|---|
| 524 | for (int_4 ith = 0; ith < map.NbThetaSlices(); ith++)
|
|---|
| 525 | {
|
|---|
| 526 | r_8 phi0;
|
|---|
| 527 | r_8 theta;
|
|---|
| 528 | // PrtTim("debut 1ere tranche ");
|
|---|
| 529 | map.GetThetaSlice(ith,theta,phi0,pixNumber ,data);
|
|---|
| 530 | phase = complex<T>((T)0.,(T)0.);
|
|---|
| 531 | double cth = cos(theta);
|
|---|
| 532 |
|
|---|
| 533 | //part of the sky out of the symetric cut
|
|---|
| 534 | bool keep_it = (fabs(cth) >= cos_theta_cut);
|
|---|
| 535 |
|
|---|
| 536 | // PrtTim("fin 1ere tranche ");
|
|---|
| 537 |
|
|---|
| 538 | if (keep_it)
|
|---|
| 539 | {
|
|---|
| 540 | // phase = CFromFourierAnalysis(nmmax,data,phi0);
|
|---|
| 541 | // PrtTim("avant Fourier ");
|
|---|
| 542 | CFromFourierAnalysis(nmmax,data,phase, phi0);
|
|---|
| 543 | // PrtTim("apres Fourier ");
|
|---|
| 544 |
|
|---|
| 545 | }
|
|---|
| 546 |
|
|---|
| 547 | // ---------------------------------------------------------------------
|
|---|
| 548 | // computes the a_lm by integrating over theta
|
|---|
| 549 | // lambda_lm(theta) * phas_m(theta)
|
|---|
| 550 | // for each m and l
|
|---|
| 551 | // -----------------------------------------------------------------------
|
|---|
| 552 | // PrtTim("avant instanciation LM ");
|
|---|
| 553 | LambdaLMBuilder lb(theta,nlmax,nmmax);
|
|---|
| 554 | // PrtTim("apres instanciation LM ");
|
|---|
| 555 | r_8 domega=map.PixSolAngle(map.PixIndexSph(theta,phi0));
|
|---|
| 556 |
|
|---|
| 557 | // PrtTim("avant mise a jour Alm ");
|
|---|
| 558 | complex<T> fi;
|
|---|
| 559 | T facteur;
|
|---|
| 560 | int index;
|
|---|
| 561 | for (int m = 0; m <= nmmax; m++)
|
|---|
| 562 | {
|
|---|
| 563 | fi = phase(m);
|
|---|
| 564 | for (int l = m; l<= nlmax; l++)
|
|---|
| 565 | {
|
|---|
| 566 | index = alm.indexOfElement(l,m);
|
|---|
| 567 | // facteur = (T)(lb.lamlm(l,m) * domega);
|
|---|
| 568 | facteur = (T)(lb.lamlm(index) * domega);
|
|---|
| 569 | // alm(l,m) += facteur * fi ;
|
|---|
| 570 | alm(index) += facteur * fi ;
|
|---|
| 571 | }
|
|---|
| 572 | }
|
|---|
| 573 |
|
|---|
| 574 |
|
|---|
| 575 |
|
|---|
| 576 | //
|
|---|
| 577 | //
|
|---|
| 578 | // PrtTim("apres mise a jour Alm ");
|
|---|
| 579 | }
|
|---|
| 580 | }
|
|---|
| 581 | /*! \fn TVector< complex<T> > SOPHYA::SphericalTransformServer::CFromFourierAnalysis(int_4 nmmax, const TVector<complex<T> >datain, r_8 phi0) const
|
|---|
| 582 |
|
|---|
| 583 | \return a vector with mmax elements which are sums :
|
|---|
| 584 | \f$\sum_{k=0}^{nphi}datain(\theta,\varphi_k)e^{im\varphi_k}\f$ for (mmax+1) values of \f$m\f$ from 0 to mmax.
|
|---|
| 585 | */
|
|---|
| 586 | template<class T>
|
|---|
| 587 | TVector< complex<T> > SphericalTransformServer<T>::CFromFourierAnalysis(int_4 nmmax, const TVector<complex<T> >datain, r_8 phi0) const
|
|---|
| 588 | {
|
|---|
| 589 | /*=======================================================================
|
|---|
| 590 | integrates (data * phi-dependence-of-Ylm) over phi
|
|---|
| 591 | --> function of m can be computed by FFT
|
|---|
| 592 |
|
|---|
| 593 | datain est modifie
|
|---|
| 594 | =======================================================================*/
|
|---|
| 595 | int_4 nph=datain.NElts();
|
|---|
| 596 | if (nph <= 0)
|
|---|
| 597 | {
|
|---|
| 598 | throw PException("bizarre : vecteur datain de longueur nulle (CFromFourierAnalysis)");
|
|---|
| 599 | }
|
|---|
| 600 | TVector<complex<T> > transformedData(nph);
|
|---|
| 601 | fftIntfPtr_-> FFTForward(datain, transformedData);
|
|---|
| 602 |
|
|---|
| 603 | TVector< complex<T> > dataout(nmmax+1);
|
|---|
| 604 |
|
|---|
| 605 | int im_max=min(nph,nmmax+1);
|
|---|
| 606 | int i;
|
|---|
| 607 | dataout = complex<T>((T)0.,(T)0.);
|
|---|
| 608 | // for (i=0;i< dataout.NElts();i++) dataout(i)=complex<T>((T)0.,(T)0.);
|
|---|
| 609 | for (i=0;i<im_max;i++) dataout(i)=transformedData(i);
|
|---|
| 610 |
|
|---|
| 611 |
|
|---|
| 612 | for (int kk=nph; kk<dataout.NElts(); kk++) dataout(kk)=dataout(kk%nph);
|
|---|
| 613 | for (i = 0;i <dataout.NElts();i++){
|
|---|
| 614 | dataout(i)*= (complex<T>)(complex<double>(cos(-i*phi0),sin(-i*phi0)));
|
|---|
| 615 | }
|
|---|
| 616 | return dataout;
|
|---|
| 617 | }
|
|---|
| 618 |
|
|---|
| 619 | //&&&&&&&&& nouvelle version
|
|---|
| 620 | /* \fn TVector< complex<T> > SOPHYA::SphericalTransformServer::CFromFourierAnalysis(int_4 nmmax, const TVector<T> datain, r_8 phi0) const
|
|---|
| 621 |
|
|---|
| 622 | same as previous one, but with a "datain" which is real (not complex) */
|
|---|
| 623 | template<class T>
|
|---|
| 624 | void SphericalTransformServer<T>::CFromFourierAnalysis(int_4 nmmax, const TVector<T> datain, TVector< complex<T> >& dataout, r_8 phi0) const
|
|---|
| 625 | {
|
|---|
| 626 | //=======================================================================
|
|---|
| 627 | // integrates (data * phi-dependence-of-Ylm) over phi
|
|---|
| 628 | // --> function of m can be computed by FFT
|
|---|
| 629 | // ! with 0<= m <= npoints/2 (: Nyquist)
|
|---|
| 630 | // ! because the data is real the negative m are the conjugate of the
|
|---|
| 631 | // ! positive ones
|
|---|
| 632 |
|
|---|
| 633 | // datain est modifie
|
|---|
| 634 | //
|
|---|
| 635 | // =======================================================================
|
|---|
| 636 | int_4 nph=datain.NElts();
|
|---|
| 637 | if (nph <= 0)
|
|---|
| 638 | {
|
|---|
| 639 | throw PException("bizarre : vecteur datain de longueur nulle (CFromFourierAnalysis)");
|
|---|
| 640 | }
|
|---|
| 641 | // if (nph%2 != 0 )
|
|---|
| 642 | // {
|
|---|
| 643 | // throw PException("SphericalTransformServer<T>::CFromFourierAnalysis : longueur de datain impair ?");
|
|---|
| 644 | // }
|
|---|
| 645 | TVector<complex<T> > transformedData;
|
|---|
| 646 |
|
|---|
| 647 | // la taille du vecteur complexe retourne est nph/2+1 (si la taille
|
|---|
| 648 | // du vecteur reel entre est nph)
|
|---|
| 649 | // cout << " longueur de datain = " << nph << endl;
|
|---|
| 650 | fftIntfPtr_-> FFTForward(datain, transformedData);
|
|---|
| 651 | // cout << " taille de la transformee " << transformedData.Size() << endl;
|
|---|
| 652 | // TVector< complex<T> > dataout(nmmax+1);
|
|---|
| 653 | dataout.ReSize(nmmax+1);
|
|---|
| 654 |
|
|---|
| 655 | // on transfere le resultat de la fft dans dataout.
|
|---|
| 656 |
|
|---|
| 657 | int maxFreqAccessiblesParFFT = min(nph/2,nmmax);
|
|---|
| 658 | int i;
|
|---|
| 659 | for (i=0;i<=maxFreqAccessiblesParFFT;i++) dataout(i)=transformedData(i);
|
|---|
| 660 |
|
|---|
| 661 |
|
|---|
| 662 | // si dataout n'est pas plein, on complete jusqu'a nph+1 valeurs (a moins
|
|---|
| 663 | // que dataout ne soit plein avant d'atteindre nph)
|
|---|
| 664 | if (maxFreqAccessiblesParFFT != nmmax )
|
|---|
| 665 | {
|
|---|
| 666 | int maxMfft = min(nph,nmmax);
|
|---|
| 667 | for (i=maxFreqAccessiblesParFFT+1; i<=maxMfft; i++)
|
|---|
| 668 | {
|
|---|
| 669 | dataout(i) = conj(dataout(nph-i) );
|
|---|
| 670 | }
|
|---|
| 671 | // on conplete, si necessaire, par periodicite
|
|---|
| 672 | if ( maxMfft != nmmax )
|
|---|
| 673 | {
|
|---|
| 674 | for (int kk=nph+1; kk <= nmmax; kk++)
|
|---|
| 675 | {
|
|---|
| 676 | dataout(kk)=dataout(kk%nph);
|
|---|
| 677 | }
|
|---|
| 678 | }
|
|---|
| 679 | }
|
|---|
| 680 | for (i = 0;i <dataout.NElts();i++)
|
|---|
| 681 | {
|
|---|
| 682 | dataout(i)*= (complex<T>)(complex<double>(cos(-i*phi0),sin(-i*phi0)));
|
|---|
| 683 | }
|
|---|
| 684 | // return dataout;
|
|---|
| 685 | }
|
|---|
| 686 |
|
|---|
| 687 | /*! \fn void SOPHYA::SphericalTransformServer::GenerateFromAlm(SphericalMap<T>& mapq,
|
|---|
| 688 | SphericalMap<T>& mapu,
|
|---|
| 689 | int_4 pixelSizeIndex,
|
|---|
| 690 | const Alm<T>& alme,
|
|---|
| 691 | const Alm<T>& almb) const
|
|---|
| 692 |
|
|---|
| 693 | synthesis of a polarization map from Alm coefficients. The spheres mapq and mapu contain respectively the Stokes parameters. */
|
|---|
| 694 | template<class T>
|
|---|
| 695 | void SphericalTransformServer<T>::GenerateFromAlm(SphericalMap<T>& mapq,
|
|---|
| 696 | SphericalMap<T>& mapu,
|
|---|
| 697 | int_4 pixelSizeIndex,
|
|---|
| 698 | const Alm<T>& alme,
|
|---|
| 699 | const Alm<T>& almb) const
|
|---|
| 700 | {
|
|---|
| 701 | /*=======================================================================
|
|---|
| 702 | computes a map form its alm for the HEALPIX pixelisation
|
|---|
| 703 | map(theta,phi) = sum_l_m a_lm Y_lm(theta,phi)
|
|---|
| 704 | = sum_m {e^(i*m*phi) sum_l a_lm*lambda_lm(theta)}
|
|---|
| 705 |
|
|---|
| 706 | where Y_lm(theta,phi) = lambda(theta) * e^(i*m*phi)
|
|---|
| 707 |
|
|---|
| 708 | * the recurrence of Ylm is the standard one (cf Num Rec)
|
|---|
| 709 | * the sum over m is done by FFT
|
|---|
| 710 |
|
|---|
| 711 | =======================================================================*/
|
|---|
| 712 | int_4 nlmax=alme.Lmax();
|
|---|
| 713 | if (nlmax != almb.Lmax())
|
|---|
| 714 | {
|
|---|
| 715 | cout << " SphericalTransformServer: les deux tableaux alm n'ont pas la meme taille" << endl;
|
|---|
| 716 | throw SzMismatchError("SphericalTransformServer: les deux tableaux alm n'ont pas la meme taille");
|
|---|
| 717 | }
|
|---|
| 718 | int_4 nmmax=nlmax;
|
|---|
| 719 | int_4 nsmax=0;
|
|---|
| 720 | mapq.Resize(pixelSizeIndex);
|
|---|
| 721 | mapu.Resize(pixelSizeIndex);
|
|---|
| 722 | string sphere_type=mapq.TypeOfMap();
|
|---|
| 723 | if (sphere_type != mapu.TypeOfMap())
|
|---|
| 724 | {
|
|---|
| 725 | cout << " SphericalTransformServer: les deux spheres ne sont pas de meme type" << endl;
|
|---|
| 726 | cout << " type 1 " << sphere_type << endl;
|
|---|
| 727 | cout << " type 2 " << mapu.TypeOfMap() << endl;
|
|---|
| 728 | throw SzMismatchError("SphericalTransformServer: les deux spheres ne sont pas de meme type");
|
|---|
| 729 |
|
|---|
| 730 | }
|
|---|
| 731 | bool healpix = true;
|
|---|
| 732 | if (sphere_type.substr(0,4) == "RING")
|
|---|
| 733 | {
|
|---|
| 734 | nsmax=mapq.SizeIndex();
|
|---|
| 735 | }
|
|---|
| 736 | else
|
|---|
| 737 | // pour une sphere Gorski le nombre de pixels est 12*nsmax**2
|
|---|
| 738 | // on calcule une quantite equivalente a nsmax pour la sphere-theta-phi
|
|---|
| 739 | // en vue de l'application du critere Healpix : nlmax<=3*nsmax-1
|
|---|
| 740 | // c'est approximatif ; a raffiner.
|
|---|
| 741 | healpix = false;
|
|---|
| 742 | if (sphere_type.substr(0,6) == "TETAFI")
|
|---|
| 743 | {
|
|---|
| 744 | nsmax=(int_4)sqrt(mapq.NbPixels()/12.);
|
|---|
| 745 | }
|
|---|
| 746 | else
|
|---|
| 747 | {
|
|---|
| 748 | cout << " unknown type of sphere : " << sphere_type << endl;
|
|---|
| 749 | throw IOExc(" unknown type of sphere ");
|
|---|
| 750 | }
|
|---|
| 751 | cout << "GenerateFromAlm: the spheres are of type : " << sphere_type << endl;
|
|---|
| 752 | cout << "GenerateFromAlm: size indices (nside) of spheres= " << nsmax << endl;
|
|---|
| 753 | cout << "GenerateFromAlm: nlmax (from Alm) = " << nlmax << endl;
|
|---|
| 754 | if (nlmax>3*nsmax-1)
|
|---|
| 755 | {
|
|---|
| 756 | cout << "GenerateFromAlm: nlmax should be <= 3*nside-1" << endl;
|
|---|
| 757 | if (sphere_type.substr(0,6) == "TETAFI")
|
|---|
| 758 | {
|
|---|
| 759 | cout << " (for this criterium, nsmax is computed as sqrt(nbPixels/12))" << endl;
|
|---|
| 760 | }
|
|---|
| 761 | }
|
|---|
| 762 | if (alme.Lmax()!=almb.Lmax())
|
|---|
| 763 | {
|
|---|
| 764 | cout << "GenerateFromAlm: arrays Alme and Almb have not the same size ? " << endl;
|
|---|
| 765 | throw SzMismatchError("SphericalTransformServer: arrays Alme and Almb have not the same size ? ");
|
|---|
| 766 | }
|
|---|
| 767 | mapFromWX(nlmax, nmmax, mapq, mapu, alme, almb, healpix);
|
|---|
| 768 | // mapFromPM(nlmax, nmmax, mapq, mapu, alme, almb);
|
|---|
| 769 | }
|
|---|
| 770 | /*! \fn void SOPHYA::SphericalTransformServer::DecomposeToAlm(const SphericalMap<T>& mapq,
|
|---|
| 771 | const SphericalMap<T>& mapu,
|
|---|
| 772 | Alm<T>& alme,
|
|---|
| 773 | Alm<T>& almb,
|
|---|
| 774 | int_4 nlmax,
|
|---|
| 775 | r_8 cos_theta_cut) const
|
|---|
| 776 |
|
|---|
| 777 | analysis of a polarization map into Alm coefficients.
|
|---|
| 778 |
|
|---|
| 779 | The spheres \c mapq and \c mapu contain respectively the Stokes parameters.
|
|---|
| 780 |
|
|---|
| 781 | \c a2lme and \c a2lmb will receive respectively electric and magnetic Alm's
|
|---|
| 782 | nlmax : maximum value of the l index
|
|---|
| 783 |
|
|---|
| 784 | \c cos_theta_cut : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 785 |
|
|---|
| 786 |
|
|---|
| 787 | */
|
|---|
| 788 | template<class T>
|
|---|
| 789 | void SphericalTransformServer<T>::DecomposeToAlm(const SphericalMap<T>& mapq,
|
|---|
| 790 | const SphericalMap<T>& mapu,
|
|---|
| 791 | Alm<T>& alme,
|
|---|
| 792 | Alm<T>& almb,
|
|---|
| 793 | int_4 nlmax,
|
|---|
| 794 | r_8 cos_theta_cut) const
|
|---|
| 795 | {
|
|---|
| 796 | DecomposeToAlm(const_cast< SphericalMap<T>& >(mapq), const_cast< SphericalMap<T>& >(mapu), alme, almb, nlmax, cos_theta_cut);
|
|---|
| 797 | }
|
|---|
| 798 |
|
|---|
| 799 | /*! \fn void SOPHYA::SphericalTransformServer::DecomposeToAlm(const SphericalMap<T>& mapq,
|
|---|
| 800 | const SphericalMap<T>& mapu,
|
|---|
| 801 | Alm<T>& alme,
|
|---|
| 802 | Alm<T>& almb,
|
|---|
| 803 | int_4 nlmax,
|
|---|
| 804 | r_8 cos_theta_cut,
|
|---|
| 805 | int iterationOrder) const
|
|---|
| 806 |
|
|---|
| 807 | analysis of a polarization map into Alm coefficients.
|
|---|
| 808 |
|
|---|
| 809 | The spheres \c mapq and \c mapu contain respectively the Stokes parameters.
|
|---|
| 810 |
|
|---|
| 811 | \c a2lme and \c a2lmb will receive respectively electric and magnetic Alm's
|
|---|
| 812 | nlmax : maximum value of the l index
|
|---|
| 813 |
|
|---|
| 814 | \c cos_theta_cut : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 815 |
|
|---|
| 816 | \param<iterationOrder> : 1,2,3,4.... order of an iterative analysis. (Default : 0 -> standard analysis). If iterationOrder is not null, the method works with SphereHEALPix but NOT WITH SphereThetaPhi maps !
|
|---|
| 817 |
|
|---|
| 818 | THE INPUT MAPS CAN BE MODIFIED (only if iterationOrder >0)
|
|---|
| 819 |
|
|---|
| 820 | */
|
|---|
| 821 | template<class T>
|
|---|
| 822 | void SphericalTransformServer<T>::DecomposeToAlm(SphericalMap<T>& mapq,
|
|---|
| 823 | SphericalMap<T>& mapu,
|
|---|
| 824 | Alm<T>& alme,
|
|---|
| 825 | Alm<T>& almb,
|
|---|
| 826 | int_4 nlmax,
|
|---|
| 827 | r_8 cos_theta_cut,
|
|---|
| 828 | int iterationOrder) const
|
|---|
| 829 | {
|
|---|
| 830 | int_4 nmmax = nlmax;
|
|---|
| 831 | carteVersAlm(mapq, mapu, alme, almb, nlmax, cos_theta_cut);
|
|---|
| 832 | if (iterationOrder > 0)
|
|---|
| 833 | {
|
|---|
| 834 | TVector<int_4> fact(iterationOrder+2);
|
|---|
| 835 | fact(0) = 1;
|
|---|
| 836 | int k;
|
|---|
| 837 | for (k=1; k <= iterationOrder+1; k++)
|
|---|
| 838 | {
|
|---|
| 839 | fact(k) = fact(k-1)*k;
|
|---|
| 840 | }
|
|---|
| 841 | Alm<T> alme2(alme);
|
|---|
| 842 | Alm<T> almb2(almb);
|
|---|
| 843 | T Tzero = (T)0.;
|
|---|
| 844 | complex<T> complexZero = complex<T>(Tzero, Tzero);
|
|---|
| 845 | alme = complexZero;
|
|---|
| 846 | almb = complexZero;
|
|---|
| 847 | int signe = 1;
|
|---|
| 848 | int nbIteration = iterationOrder+1;
|
|---|
| 849 | for (k=1; k <= nbIteration; k++)
|
|---|
| 850 | {
|
|---|
| 851 | T facMult = (T)(0.5*signe*fact(iterationOrder)*(2*nbIteration-k)/(fact(k)*fact(nbIteration-k)));
|
|---|
| 852 | for (int m = 0; m <= nmmax; m++)
|
|---|
| 853 | {
|
|---|
| 854 | for (int l = m; l<= nlmax; l++)
|
|---|
| 855 | {
|
|---|
| 856 | alme(l,m) += facMult*alme2(l,m);
|
|---|
| 857 | almb(l,m) += facMult*almb2(l,m);
|
|---|
| 858 | }
|
|---|
| 859 | }
|
|---|
| 860 | if (k == nbIteration) break;
|
|---|
| 861 | signe = -signe;
|
|---|
| 862 | for (int k=0; k< mapq.NbPixels(); k++)
|
|---|
| 863 | {
|
|---|
| 864 | mapq(k) = (T)0.;
|
|---|
| 865 | mapu(k) = (T)0.;
|
|---|
| 866 | }
|
|---|
| 867 | // synthetize a map from the estimated alm
|
|---|
| 868 | GenerateFromAlm(mapq,mapu,mapq.SizeIndex(),alme2,almb2);
|
|---|
| 869 | alme2 = complexZero;
|
|---|
| 870 | almb2 = complexZero;
|
|---|
| 871 | // analyse the new map
|
|---|
| 872 | carteVersAlm(mapq, mapu, alme2, almb2, nlmax, cos_theta_cut);
|
|---|
| 873 | }
|
|---|
| 874 | }
|
|---|
| 875 | }
|
|---|
| 876 |
|
|---|
| 877 | template<class T>
|
|---|
| 878 | void SphericalTransformServer<T>::carteVersAlm(const SphericalMap<T>& mapq,
|
|---|
| 879 | const SphericalMap<T>& mapu,
|
|---|
| 880 | Alm<T>& alme,
|
|---|
| 881 | Alm<T>& almb,
|
|---|
| 882 | int_4 nlmax,
|
|---|
| 883 | r_8 cos_theta_cut) const
|
|---|
| 884 | {
|
|---|
| 885 | int_4 nmmax = nlmax;
|
|---|
| 886 | // resize et remise a zero
|
|---|
| 887 | alme.ReSizeToLmax(nlmax);
|
|---|
| 888 | almb.ReSizeToLmax(nlmax);
|
|---|
| 889 |
|
|---|
| 890 |
|
|---|
| 891 | TVector<T> dataq;
|
|---|
| 892 | TVector<T> datau;
|
|---|
| 893 | TVector<int_4> pixNumber;
|
|---|
| 894 |
|
|---|
| 895 | string sphere_type=mapq.TypeOfMap();
|
|---|
| 896 | if (sphere_type != mapu.TypeOfMap())
|
|---|
| 897 | {
|
|---|
| 898 | cout << " SphericalTransformServer: les deux spheres ne sont pas de meme type" << endl;
|
|---|
| 899 | cout << " type 1 " << sphere_type << endl;
|
|---|
| 900 | cout << " type 2 " << mapu.TypeOfMap() << endl;
|
|---|
| 901 | throw SzMismatchError("SphericalTransformServer: les deux spheres ne sont pas de meme type");
|
|---|
| 902 |
|
|---|
| 903 | }
|
|---|
| 904 | if (mapq.NbPixels()!=mapu.NbPixels())
|
|---|
| 905 | {
|
|---|
| 906 | cout << " DecomposeToAlm: map Q and map U have not same size ?" << endl;
|
|---|
| 907 | throw SzMismatchError("SphericalTransformServer::DecomposeToAlm: map Q and map U have not same size ");
|
|---|
| 908 | }
|
|---|
| 909 | for (int_4 ith = 0; ith < mapq.NbThetaSlices(); ith++)
|
|---|
| 910 | {
|
|---|
| 911 | r_8 phi0;
|
|---|
| 912 | r_8 theta;
|
|---|
| 913 | mapq.GetThetaSlice(ith,theta,phi0, pixNumber,dataq);
|
|---|
| 914 | mapu.GetThetaSlice(ith,theta,phi0, pixNumber,datau);
|
|---|
| 915 | if (dataq.NElts() != datau.NElts() )
|
|---|
| 916 | {
|
|---|
| 917 | throw SzMismatchError("the spheres have not the same pixelization");
|
|---|
| 918 | }
|
|---|
| 919 | r_8 domega=mapq.PixSolAngle(mapq.PixIndexSph(theta,phi0));
|
|---|
| 920 | double cth = cos(theta);
|
|---|
| 921 | //part of the sky out of the symetric cut
|
|---|
| 922 | bool keep_it = (fabs(cth) >= cos_theta_cut);
|
|---|
| 923 | if (keep_it)
|
|---|
| 924 | {
|
|---|
| 925 | // almFromPM(pixNumber.NElts(), nlmax, nmmax, phi0, domega, theta, dataq, datau, alme, almb);
|
|---|
| 926 | almFromWX(nlmax, nmmax, phi0, domega, theta, dataq, datau, alme, almb);
|
|---|
| 927 | }
|
|---|
| 928 | }
|
|---|
| 929 | }
|
|---|
| 930 |
|
|---|
| 931 |
|
|---|
| 932 | /*! \fn void SOPHYA::SphericalTransformServer::almFromWX(int_4 nlmax, int_4 nmmax,
|
|---|
| 933 | r_8 phi0, r_8 domega,
|
|---|
| 934 | r_8 theta,
|
|---|
| 935 | const TVector<T>& dataq,
|
|---|
| 936 | const TVector<T>& datau,
|
|---|
| 937 | Alm<T>& alme,
|
|---|
| 938 | Alm<T>& almb) const
|
|---|
| 939 |
|
|---|
| 940 | Compute polarized Alm's as :
|
|---|
| 941 | \f[
|
|---|
| 942 | a_{lm}^E=\frac{1}{\sqrt{2}}\sum_{slices}{\omega_{pix}\left(\,_{w}\lambda_l^m\tilde{Q}-i\,_{x}\lambda_l^m\tilde{U}\right)}
|
|---|
| 943 | \f]
|
|---|
| 944 | \f[
|
|---|
| 945 | a_{lm}^B=\frac{1}{\sqrt{2}}\sum_{slices}{\omega_{pix}\left(i\,_{x}\lambda_l^m\tilde{Q}+\,_{w}\lambda_l^m\tilde{U}\right)}
|
|---|
| 946 | \f]
|
|---|
| 947 |
|
|---|
| 948 | where \f$\tilde{Q}\f$ and \f$\tilde{U}\f$ are C-coefficients computed by FFT (method CFromFourierAnalysis, called by present method) from the Stokes parameters.
|
|---|
| 949 |
|
|---|
| 950 | \f$\omega_{pix}\f$ are solid angle of each pixel.
|
|---|
| 951 |
|
|---|
| 952 | dataq, datau : Stokes parameters.
|
|---|
| 953 |
|
|---|
| 954 | */
|
|---|
| 955 | template<class T>
|
|---|
| 956 | void SphericalTransformServer<T>::almFromWX(int_4 nlmax, int_4 nmmax,
|
|---|
| 957 | r_8 phi0, r_8 domega,
|
|---|
| 958 | r_8 theta,
|
|---|
| 959 | const TVector<T>& dataq,
|
|---|
| 960 | const TVector<T>& datau,
|
|---|
| 961 | Alm<T>& alme,
|
|---|
| 962 | Alm<T>& almb) const
|
|---|
| 963 | {
|
|---|
| 964 | TVector< complex<T> > phaseq(nmmax+1);
|
|---|
| 965 | TVector< complex<T> > phaseu(nmmax+1);
|
|---|
| 966 | // TVector<complex<T> > datain(nph);
|
|---|
| 967 | for (int i=0;i< nmmax+1;i++)
|
|---|
| 968 | {
|
|---|
| 969 | phaseq(i)=0;
|
|---|
| 970 | phaseu(i)=0;
|
|---|
| 971 | }
|
|---|
| 972 | // for(int kk=0; kk<nph; kk++) datain(kk)=complex<T>(dataq(kk),0.);
|
|---|
| 973 |
|
|---|
| 974 | // phaseq = CFromFourierAnalysis(nmmax,dataq,phi0);
|
|---|
| 975 | CFromFourierAnalysis(nmmax,dataq,phaseq, phi0);
|
|---|
| 976 |
|
|---|
| 977 | // phaseu= CFromFourierAnalysis(nmmax,datau,phi0);
|
|---|
| 978 | CFromFourierAnalysis(nmmax,datau,phaseu, phi0);
|
|---|
| 979 |
|
|---|
| 980 | LambdaWXBuilder lwxb(theta,nlmax,nmmax);
|
|---|
| 981 |
|
|---|
| 982 | r_8 sqr2inv=1/Rac2;
|
|---|
| 983 | for (int m = 0; m <= nmmax; m++)
|
|---|
| 984 | {
|
|---|
| 985 | r_8 lambda_w=0.;
|
|---|
| 986 | r_8 lambda_x=0.;
|
|---|
| 987 | lwxb.lam_wx(m, m, lambda_w, lambda_x);
|
|---|
| 988 | complex<T> zi_lam_x((T)0., (T)lambda_x);
|
|---|
| 989 | alme(m,m) += ( (T)(lambda_w)*phaseq(m)-zi_lam_x*phaseu(m) )*(T)(domega*sqr2inv);
|
|---|
| 990 | almb(m,m) += ( (T)(lambda_w)*phaseu(m)+zi_lam_x*phaseq(m) )*(T)(domega*sqr2inv);
|
|---|
| 991 |
|
|---|
| 992 | for (int l = m+1; l<= nlmax; l++)
|
|---|
| 993 | {
|
|---|
| 994 | lwxb.lam_wx(l, m, lambda_w, lambda_x);
|
|---|
| 995 | zi_lam_x = complex<T>((T)0., (T)lambda_x);
|
|---|
| 996 | alme(l,m) += ( (T)(lambda_w)*phaseq(m)-zi_lam_x*phaseu(m) )*(T)(domega*sqr2inv);
|
|---|
| 997 | almb(l,m) += ( (T)(lambda_w)*phaseu(m)+zi_lam_x*phaseq(m) )*(T)(domega*sqr2inv);
|
|---|
| 998 | }
|
|---|
| 999 | }
|
|---|
| 1000 | }
|
|---|
| 1001 |
|
|---|
| 1002 |
|
|---|
| 1003 | /*! \fn void SOPHYA::SphericalTransformServer::almFromPM(int_4 nph, int_4 nlmax,
|
|---|
| 1004 | int_4 nmmax,
|
|---|
| 1005 | r_8 phi0, r_8 domega,
|
|---|
| 1006 | r_8 theta,
|
|---|
| 1007 | const TVector<T>& dataq,
|
|---|
| 1008 | const TVector<T>& datau,
|
|---|
| 1009 | Alm<T>& alme,
|
|---|
| 1010 | Alm<T>& almb) const
|
|---|
| 1011 |
|
|---|
| 1012 | Compute polarized Alm's as :
|
|---|
| 1013 | \f[
|
|---|
| 1014 | a_{lm}^E=-\frac{1}{2}\sum_{slices}{\omega_{pix}\left(\,_{+}\lambda_l^m\tilde{P^+}+\,_{-}\lambda_l^m\tilde{P^-}\right)}
|
|---|
| 1015 | \f]
|
|---|
| 1016 | \f[
|
|---|
| 1017 | a_{lm}^B=\frac{i}{2}\sum_{slices}{\omega_{pix}\left(\,_{+}\lambda_l^m\tilde{P^+}-\,_{-}\lambda_l^m\tilde{P^-}\right)}
|
|---|
| 1018 | \f]
|
|---|
| 1019 |
|
|---|
| 1020 | where \f$\tilde{P^{\pm}}=\tilde{Q}\pm\tilde{U}\f$ computed by FFT (method CFromFourierAnalysis, called by present method) from the Stokes parameters,\f$Q\f$ and \f$U\f$ .
|
|---|
| 1021 |
|
|---|
| 1022 | \f$\omega_{pix}\f$ are solid angle of each pixel.
|
|---|
| 1023 |
|
|---|
| 1024 | dataq, datau : Stokes parameters.
|
|---|
| 1025 |
|
|---|
| 1026 | */
|
|---|
| 1027 | template<class T>
|
|---|
| 1028 | void SphericalTransformServer<T>::almFromPM(int_4 nph, int_4 nlmax,
|
|---|
| 1029 | int_4 nmmax,
|
|---|
| 1030 | r_8 phi0, r_8 domega,
|
|---|
| 1031 | r_8 theta,
|
|---|
| 1032 | const TVector<T>& dataq,
|
|---|
| 1033 | const TVector<T>& datau,
|
|---|
| 1034 | Alm<T>& alme,
|
|---|
| 1035 | Alm<T>& almb) const
|
|---|
| 1036 | {
|
|---|
| 1037 | TVector< complex<T> > phasep(nmmax+1);
|
|---|
| 1038 | TVector< complex<T> > phasem(nmmax+1);
|
|---|
| 1039 | TVector<complex<T> > datain(nph);
|
|---|
| 1040 | for (int i=0;i< nmmax+1;i++)
|
|---|
| 1041 | {
|
|---|
| 1042 | phasep(i)=0;
|
|---|
| 1043 | phasem(i)=0;
|
|---|
| 1044 | }
|
|---|
| 1045 | int kk;
|
|---|
| 1046 | for(kk=0; kk<nph; kk++) datain(kk)=complex<T>(dataq(kk),datau(kk));
|
|---|
| 1047 |
|
|---|
| 1048 | phasep = CFromFourierAnalysis(nmmax,datain,phi0);
|
|---|
| 1049 |
|
|---|
| 1050 | for(kk=0; kk<nph; kk++) datain(kk)=complex<T>(dataq(kk),-datau(kk));
|
|---|
| 1051 | phasem = CFromFourierAnalysis(nmmax,datain,phi0);
|
|---|
| 1052 | LambdaPMBuilder lpmb(theta,nlmax,nmmax);
|
|---|
| 1053 |
|
|---|
| 1054 | for (int m = 0; m <= nmmax; m++)
|
|---|
| 1055 | {
|
|---|
| 1056 | r_8 lambda_p=0.;
|
|---|
| 1057 | r_8 lambda_m=0.;
|
|---|
| 1058 | complex<T> im((T)0.,(T)1.);
|
|---|
| 1059 | lpmb.lam_pm(m, m, lambda_p, lambda_m);
|
|---|
| 1060 |
|
|---|
| 1061 | alme(m,m) += -( (T)(lambda_p)*phasep(m) + (T)(lambda_m)*phasem(m) )*(T)(domega*0.5);
|
|---|
| 1062 | almb(m,m) += im*( (T)(lambda_p)*phasep(m) - (T)(lambda_m)*phasem(m) )*(T)(domega*0.5);
|
|---|
| 1063 | for (int l = m+1; l<= nlmax; l++)
|
|---|
| 1064 | {
|
|---|
| 1065 | lpmb.lam_pm(l, m, lambda_p, lambda_m);
|
|---|
| 1066 | alme(l,m) += -( (T)(lambda_p)*phasep(m) + (T)(lambda_m)*phasem(m) )*(T)(domega*0.5);
|
|---|
| 1067 | almb(l,m) += im* ( (T)(lambda_p)*phasep(m) - (T)(lambda_m)*phasem(m) )*(T)(domega*0.5);
|
|---|
| 1068 | }
|
|---|
| 1069 | }
|
|---|
| 1070 | }
|
|---|
| 1071 |
|
|---|
| 1072 |
|
|---|
| 1073 | /*! \fn void SOPHYA::SphericalTransformServer::mapFromWX(int_4 nlmax, int_4 nmmax,
|
|---|
| 1074 | SphericalMap<T>& mapq,
|
|---|
| 1075 | SphericalMap<T>& mapu,
|
|---|
| 1076 | const Alm<T>& alme,
|
|---|
| 1077 | const Alm<T>& almb, bool healpix) const
|
|---|
| 1078 |
|
|---|
| 1079 | synthesis of Stokes parameters following formulae :
|
|---|
| 1080 |
|
|---|
| 1081 | \f[
|
|---|
| 1082 | Q=\sum_{m=-mmax}^{mmax}b_m^qe^{im\varphi}
|
|---|
| 1083 | \f]
|
|---|
| 1084 | \f[
|
|---|
| 1085 | U=\sum_{m=-mmax}^{mmax}b_m^ue^{im\varphi}
|
|---|
| 1086 | \f]
|
|---|
| 1087 |
|
|---|
| 1088 | computed by FFT (method fourierSynthesisFromB called by the present one)
|
|---|
| 1089 |
|
|---|
| 1090 | with :
|
|---|
| 1091 |
|
|---|
| 1092 | \f[
|
|---|
| 1093 | b_m^q=-\frac{1}{\sqrt{2}}\sum_{l=|m|}^{lmax}{\left(\,_{w}\lambda_l^ma_{lm}^E-i\,_{x}\lambda_l^ma_{lm}^B\right) }
|
|---|
| 1094 | \f]
|
|---|
| 1095 | \f[
|
|---|
| 1096 | b_m^u=\frac{1}{\sqrt{2}}\sum_{l=|m|}^{lmax}{\left(i\,_{x}\lambda_l^ma_{lm}^E+\,_{w}\lambda_l^ma_{lm}^B\right) }
|
|---|
| 1097 | \f]
|
|---|
| 1098 | */
|
|---|
| 1099 | template<class T>
|
|---|
| 1100 | void SphericalTransformServer<T>::mapFromWX(int_4 nlmax, int_4 nmmax,
|
|---|
| 1101 | SphericalMap<T>& mapq,
|
|---|
| 1102 | SphericalMap<T>& mapu,
|
|---|
| 1103 | const Alm<T>& alme,
|
|---|
| 1104 | const Alm<T>& almb, bool healpix) const
|
|---|
| 1105 | {
|
|---|
| 1106 | int i;
|
|---|
| 1107 |
|
|---|
| 1108 | Bm<complex<T> > b_m_theta_q(nmmax);
|
|---|
| 1109 | Bm<complex<T> > b_m_theta_u(nmmax);
|
|---|
| 1110 |
|
|---|
| 1111 | for (int_4 ith = 0; ith < mapq.NbThetaSlices();ith++)
|
|---|
| 1112 | {
|
|---|
| 1113 | int_4 nph;
|
|---|
| 1114 | r_8 phi0;
|
|---|
| 1115 | r_8 theta;
|
|---|
| 1116 | TVector<int_4> pixNumber;
|
|---|
| 1117 | TVector<T> datan;
|
|---|
| 1118 |
|
|---|
| 1119 | mapq.GetThetaSlice(ith,theta,phi0, pixNumber,datan);
|
|---|
| 1120 | nph = pixNumber.NElts();
|
|---|
| 1121 | // -----------------------------------------------------
|
|---|
| 1122 | // for each theta, and each m, computes
|
|---|
| 1123 | // b(m,theta) = sum_over_l>m (lambda_l_m(theta) * a_l_m)
|
|---|
| 1124 | // ------------------------------------------------------
|
|---|
| 1125 | LambdaWXBuilder lwxb(theta,nlmax,nmmax);
|
|---|
| 1126 | // LambdaPMBuilder lpmb(theta,nlmax,nmmax);
|
|---|
| 1127 | r_8 sqr2inv=1/Rac2;
|
|---|
| 1128 | int m;
|
|---|
| 1129 | for (m = 0; m <= nmmax; m++)
|
|---|
| 1130 | {
|
|---|
| 1131 | r_8 lambda_w=0.;
|
|---|
| 1132 | r_8 lambda_x=0.;
|
|---|
| 1133 | lwxb.lam_wx(m, m, lambda_w, lambda_x);
|
|---|
| 1134 | complex<T> zi_lam_x((T)0., (T)lambda_x);
|
|---|
| 1135 |
|
|---|
| 1136 | b_m_theta_q(m) = ( (T)(lambda_w) * alme(m,m) - zi_lam_x * almb(m,m))*(T)sqr2inv ;
|
|---|
| 1137 | b_m_theta_u(m) = ( (T)(lambda_w) * almb(m,m) + zi_lam_x * alme(m,m))*(T)sqr2inv;
|
|---|
| 1138 |
|
|---|
| 1139 |
|
|---|
| 1140 | for (int l = m+1; l<= nlmax; l++)
|
|---|
| 1141 | {
|
|---|
| 1142 |
|
|---|
| 1143 | lwxb.lam_wx(l, m, lambda_w, lambda_x);
|
|---|
| 1144 | zi_lam_x= complex<T>((T)0., (T)lambda_x);
|
|---|
| 1145 |
|
|---|
| 1146 | b_m_theta_q(m) += ((T)(lambda_w)*alme(l,m)-zi_lam_x *almb(l,m))*(T)sqr2inv;
|
|---|
| 1147 | b_m_theta_u(m) += ((T)(lambda_w)*almb(l,m)+zi_lam_x *alme(l,m))*(T)sqr2inv;
|
|---|
| 1148 |
|
|---|
| 1149 | }
|
|---|
| 1150 | }
|
|---|
| 1151 | // obtains the negative m of b(m,theta) (= complex conjugate)
|
|---|
| 1152 | for (m=1;m<=nmmax;m++)
|
|---|
| 1153 | {
|
|---|
| 1154 | b_m_theta_q(-m) = conj(b_m_theta_q(m));
|
|---|
| 1155 | b_m_theta_u(-m) = conj(b_m_theta_u(m));
|
|---|
| 1156 | }
|
|---|
| 1157 | if (healpix)
|
|---|
| 1158 | {
|
|---|
| 1159 | TVector<T> Tempq = RfourierSynthesisFromB(b_m_theta_q,nph,phi0);
|
|---|
| 1160 | TVector<T> Tempu = RfourierSynthesisFromB(b_m_theta_u,nph,phi0);
|
|---|
| 1161 | for (i=0;i< nph;i++)
|
|---|
| 1162 | {
|
|---|
| 1163 | mapq(pixNumber(i))=Tempq(i);
|
|---|
| 1164 | mapu(pixNumber(i))=Tempu(i);
|
|---|
| 1165 | }
|
|---|
| 1166 | }
|
|---|
| 1167 | else
|
|---|
| 1168 | // pour des pixelisations quelconques (autres que HEALPix
|
|---|
| 1169 | // nph n'est pas toujours pair
|
|---|
| 1170 | // ca fait des problemes pour les transformees de Fourier
|
|---|
| 1171 | // car le server de TF ajuste la longueur du vecteur reel
|
|---|
| 1172 | // en sortie de TF, bref, la securite veut qu'on prenne une
|
|---|
| 1173 | // TF complexe
|
|---|
| 1174 | {
|
|---|
| 1175 | TVector<complex<T> > Tempq = fourierSynthesisFromB(b_m_theta_q,nph,phi0);
|
|---|
| 1176 | TVector<complex<T> > Tempu = fourierSynthesisFromB(b_m_theta_u,nph,phi0);
|
|---|
| 1177 | for (i=0;i< nph;i++)
|
|---|
| 1178 | {
|
|---|
| 1179 | mapq(pixNumber(i))=Tempq(i).real();
|
|---|
| 1180 | mapu(pixNumber(i))=Tempu(i).real();
|
|---|
| 1181 | }
|
|---|
| 1182 | }
|
|---|
| 1183 | }
|
|---|
| 1184 | }
|
|---|
| 1185 | /*! \fn void SOPHYA::SphericalTransformServer::mapFromPM(int_4 nlmax, int_4 nmmax,
|
|---|
| 1186 | SphericalMap<T>& mapq,
|
|---|
| 1187 | SphericalMap<T>& mapu,
|
|---|
| 1188 | const Alm<T>& alme,
|
|---|
| 1189 | const Alm<T>& almb) const
|
|---|
| 1190 |
|
|---|
| 1191 | synthesis of polarizations following formulae :
|
|---|
| 1192 |
|
|---|
| 1193 | \f[
|
|---|
| 1194 | P^+ = \sum_{m=-mmax}^{mmax} {b_m^+e^{im\varphi} }
|
|---|
| 1195 | \f]
|
|---|
| 1196 | \f[
|
|---|
| 1197 | P^- = \sum_{m=-mmax}^{mmax} {b_m^-e^{im\varphi} }
|
|---|
| 1198 | \f]
|
|---|
| 1199 |
|
|---|
| 1200 | computed by FFT (method fourierSynthesisFromB called by the present one)
|
|---|
| 1201 |
|
|---|
| 1202 | with :
|
|---|
| 1203 |
|
|---|
| 1204 | \f[
|
|---|
| 1205 | b_m^+=-\sum_{l=|m|}^{lmax}{\,_{+}\lambda_l^m \left( a_{lm}^E+ia_{lm}^B \right) }
|
|---|
| 1206 | \f]
|
|---|
| 1207 | \f[
|
|---|
| 1208 | b_m^-=-\sum_{l=|m|}^{lmax}{\,_{+}\lambda_l^m \left( a_{lm}^E-ia_{lm}^B \right) }
|
|---|
| 1209 | \f]
|
|---|
| 1210 | */
|
|---|
| 1211 | template<class T>
|
|---|
| 1212 | void SphericalTransformServer<T>::mapFromPM(int_4 nlmax, int_4 nmmax,
|
|---|
| 1213 | SphericalMap<T>& mapq,
|
|---|
| 1214 | SphericalMap<T>& mapu,
|
|---|
| 1215 | const Alm<T>& alme,
|
|---|
| 1216 | const Alm<T>& almb) const
|
|---|
| 1217 | {
|
|---|
| 1218 | Bm<complex<T> > b_m_theta_p(nmmax);
|
|---|
| 1219 | Bm<complex<T> > b_m_theta_m(nmmax);
|
|---|
| 1220 | for (int_4 ith = 0; ith < mapq.NbThetaSlices();ith++)
|
|---|
| 1221 | {
|
|---|
| 1222 | int_4 nph;
|
|---|
| 1223 | r_8 phi0;
|
|---|
| 1224 | r_8 theta;
|
|---|
| 1225 | TVector<int_4> pixNumber;
|
|---|
| 1226 | TVector<T> datan;
|
|---|
| 1227 |
|
|---|
| 1228 | mapq.GetThetaSlice(ith,theta,phi0, pixNumber,datan);
|
|---|
| 1229 | nph = pixNumber.NElts();
|
|---|
| 1230 |
|
|---|
| 1231 | // -----------------------------------------------------
|
|---|
| 1232 | // for each theta, and each m, computes
|
|---|
| 1233 | // b(m,theta) = sum_over_l>m (lambda_l_m(theta) * a_l_m)
|
|---|
| 1234 | //------------------------------------------------------
|
|---|
| 1235 |
|
|---|
| 1236 | LambdaPMBuilder lpmb(theta,nlmax,nmmax);
|
|---|
| 1237 | int m;
|
|---|
| 1238 | for (m = 0; m <= nmmax; m++)
|
|---|
| 1239 | {
|
|---|
| 1240 | r_8 lambda_p=0.;
|
|---|
| 1241 | r_8 lambda_m=0.;
|
|---|
| 1242 | lpmb.lam_pm(m, m, lambda_p, lambda_m);
|
|---|
| 1243 | complex<T> im((T)0.,(T)1.);
|
|---|
| 1244 |
|
|---|
| 1245 | b_m_theta_p(m) = (T)(lambda_p )* (-alme(m,m) - im * almb(m,m));
|
|---|
| 1246 | b_m_theta_m(m) = (T)(lambda_m) * (-alme(m,m) + im * almb(m,m));
|
|---|
| 1247 |
|
|---|
| 1248 |
|
|---|
| 1249 | for (int l = m+1; l<= nlmax; l++)
|
|---|
| 1250 | {
|
|---|
| 1251 | lpmb.lam_pm(l, m, lambda_p, lambda_m);
|
|---|
| 1252 | b_m_theta_p(m) += (T)(lambda_p)*(-alme(l,m)-im *almb(l,m));
|
|---|
| 1253 | b_m_theta_m(m) += (T)(lambda_m)*(-alme(l,m)+im *almb(l,m));
|
|---|
| 1254 | }
|
|---|
| 1255 | }
|
|---|
| 1256 |
|
|---|
| 1257 | // obtains the negative m of b(m,theta) (= complex conjugate)
|
|---|
| 1258 | for (m=1;m<=nmmax;m++)
|
|---|
| 1259 | {
|
|---|
| 1260 | b_m_theta_p(-m) = conj(b_m_theta_m(m));
|
|---|
| 1261 | b_m_theta_m(-m) = conj(b_m_theta_p(m));
|
|---|
| 1262 | }
|
|---|
| 1263 |
|
|---|
| 1264 | TVector<complex<T> > Tempp = fourierSynthesisFromB(b_m_theta_p,nph,phi0);
|
|---|
| 1265 | TVector<complex<T> > Tempm = fourierSynthesisFromB(b_m_theta_m,nph,phi0);
|
|---|
| 1266 |
|
|---|
| 1267 | for (int i=0;i< nph;i++)
|
|---|
| 1268 | {
|
|---|
| 1269 | mapq(pixNumber(i))=0.5*(Tempp(i)+Tempm(i)).real();
|
|---|
| 1270 | mapu(pixNumber(i))=0.5*(Tempp(i)-Tempm(i)).imag();
|
|---|
| 1271 | }
|
|---|
| 1272 | }
|
|---|
| 1273 | }
|
|---|
| 1274 |
|
|---|
| 1275 |
|
|---|
| 1276 | /*! \fn void SOPHYA::SphericalTransformServer::GenerateFromCl(SphericalMap<T>& sphq,
|
|---|
| 1277 | SphericalMap<T>& sphu,
|
|---|
| 1278 | int_4 pixelSizeIndex,
|
|---|
| 1279 | const TVector<T>& Cle,
|
|---|
| 1280 | const TVector<T>& Clb,
|
|---|
| 1281 | const r_8 fwhm) const
|
|---|
| 1282 |
|
|---|
| 1283 | synthesis of a polarization map from power spectra electric-Cl and magnetic-Cl (Alm's are generated randomly, following a gaussian distribution).
|
|---|
| 1284 | \param fwhm FWHM in arcmin for random generation of Alm's (eg. 5)
|
|---|
| 1285 | */
|
|---|
| 1286 | template<class T>
|
|---|
| 1287 | void SphericalTransformServer<T>::GenerateFromCl(SphericalMap<T>& sphq,
|
|---|
| 1288 | SphericalMap<T>& sphu,
|
|---|
| 1289 | int_4 pixelSizeIndex,
|
|---|
| 1290 | const TVector<T>& Cle,
|
|---|
| 1291 | const TVector<T>& Clb,
|
|---|
| 1292 | const r_8 fwhm) const
|
|---|
| 1293 | {
|
|---|
| 1294 | if (Cle.NElts() != Clb.NElts())
|
|---|
| 1295 | {
|
|---|
| 1296 | cout << " SphericalTransformServer: les deux tableaux Cl n'ont pas la meme taille" << endl;
|
|---|
| 1297 | throw SzMismatchError("SphericalTransformServer::GenerateFromCl : two Cl arrays have not same size");
|
|---|
| 1298 | }
|
|---|
| 1299 |
|
|---|
| 1300 | // Alm<T> a2lme,a2lmb;
|
|---|
| 1301 | // almFromCl(a2lme, Cle, fwhm);
|
|---|
| 1302 | // almFromCl(a2lmb, Clb, fwhm);
|
|---|
| 1303 | // Alm<T> a2lme = almFromCl(Cle, fwhm);
|
|---|
| 1304 | // Alm<T> a2lmb = almFromCl(Clb, fwhm);
|
|---|
| 1305 | Alm<T> a2lme(Cle, fwhm);
|
|---|
| 1306 | Alm<T> a2lmb(Clb, fwhm);
|
|---|
| 1307 |
|
|---|
| 1308 | GenerateFromAlm(sphq,sphu,pixelSizeIndex,a2lme,a2lmb);
|
|---|
| 1309 | }
|
|---|
| 1310 | /*! \fn void SOPHYA::SphericalTransformServer::GenerateFromCl(SphericalMap<T>& sph,
|
|---|
| 1311 | int_4 pixelSizeIndex,
|
|---|
| 1312 | const TVector<T>& Cl,
|
|---|
| 1313 | const r_8 fwhm) const
|
|---|
| 1314 |
|
|---|
| 1315 | synthesis of a temperature map from power spectrum Cl (Alm's are generated randomly, following a gaussian distribution). */
|
|---|
| 1316 | template<class T>
|
|---|
| 1317 | void SphericalTransformServer<T>::GenerateFromCl(SphericalMap<T>& sph,
|
|---|
| 1318 | int_4 pixelSizeIndex,
|
|---|
| 1319 | const TVector<T>& Cl,
|
|---|
| 1320 | const r_8 fwhm) const
|
|---|
| 1321 | {
|
|---|
| 1322 |
|
|---|
| 1323 | Alm<T> alm(Cl, fwhm);
|
|---|
| 1324 | GenerateFromAlm(sph,pixelSizeIndex, alm );
|
|---|
| 1325 | }
|
|---|
| 1326 |
|
|---|
| 1327 |
|
|---|
| 1328 |
|
|---|
| 1329 | /*! \fn TVector<T> SOPHYA::SphericalTransformServer::DecomposeToCl(SphericalMap<T>& sph, int_4 nlmax, r_8 cos_theta_cut, int iterationOrder) const
|
|---|
| 1330 |
|
|---|
| 1331 | \return power spectrum from analysis of a temperature map. THE MAP CAN BE MODIFIED (if iterationOrder >0)
|
|---|
| 1332 |
|
|---|
| 1333 | \param<nlmax> : maximum value of the l index
|
|---|
| 1334 |
|
|---|
| 1335 | \param<cos_theta_cut> : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 1336 |
|
|---|
| 1337 | \param<iterationOrder> : 1,2,3,4.... order of an iterative analysis. If iterationOrder is not null, the method works with SphereHEALPix but NOT WITH SphereThetaPhi maps !
|
|---|
| 1338 |
|
|---|
| 1339 | */
|
|---|
| 1340 | template <class T>
|
|---|
| 1341 | TVector<T> SphericalTransformServer<T>::DecomposeToCl(SphericalMap<T>& sph, int_4 nlmax, r_8 cos_theta_cut, int iterationOrder) const
|
|---|
| 1342 | {
|
|---|
| 1343 | Alm<T> alm;
|
|---|
| 1344 | DecomposeToAlm( sph, alm, nlmax, cos_theta_cut, iterationOrder);
|
|---|
| 1345 | // power spectrum
|
|---|
| 1346 | return alm.powerSpectrum();
|
|---|
| 1347 | }
|
|---|
| 1348 |
|
|---|
| 1349 |
|
|---|
| 1350 | /*! \fn TVector<T> SOPHYA::SphericalTransformServer::DecomposeToCl(const SphericalMap<T>& sph, int_4 nlmax, r_8 cos_theta_cut) const
|
|---|
| 1351 |
|
|---|
| 1352 | \return power spectrum from analysis of a temperature map.
|
|---|
| 1353 |
|
|---|
| 1354 | \param<nlmax> : maximum value of the l index
|
|---|
| 1355 |
|
|---|
| 1356 | \param<cos_theta_cut> : cosinus of the symmetric cut EULER angle theta : cos_theta_cut=0 means no cut ; cos_theta_cut=1 all the sphere is cut.
|
|---|
| 1357 |
|
|---|
| 1358 |
|
|---|
| 1359 | */
|
|---|
| 1360 |
|
|---|
| 1361 |
|
|---|
| 1362 | template <class T>
|
|---|
| 1363 | TVector<T> SphericalTransformServer<T>::DecomposeToCl(const SphericalMap<T>& sph, int_4 nlmax, r_8 cos_theta_cut) const
|
|---|
| 1364 | {
|
|---|
| 1365 | Alm<T> alm;
|
|---|
| 1366 | DecomposeToAlm( sph, alm, nlmax, cos_theta_cut);
|
|---|
| 1367 | // power spectrum
|
|---|
| 1368 | return alm.powerSpectrum();
|
|---|
| 1369 | }
|
|---|
| 1370 |
|
|---|
| 1371 | #ifdef __CXX_PRAGMA_TEMPLATES__
|
|---|
| 1372 | #pragma define_template SphericalTransformServer<r_8>
|
|---|
| 1373 | #pragma define_template SphericalTransformServer<r_4>
|
|---|
| 1374 | #endif
|
|---|
| 1375 | #if defined(ANSI_TEMPLATES) || defined(GNU_TEMPLATES)
|
|---|
| 1376 | template class SphericalTransformServer<r_8>;
|
|---|
| 1377 | template class SphericalTransformServer<r_4>;
|
|---|
| 1378 | #endif
|
|---|