00001
00002
00003
00004
00005
00006
00007
00008
00009
00010
00011 #ifndef EIGEN_REAL_SCHUR_H
00012 #define EIGEN_REAL_SCHUR_H
00013
00014 #include "./HessenbergDecomposition.h"
00015
00016 namespace Eigen {
00017
00054 template<typename _MatrixType> class RealSchur
00055 {
00056 public:
00057 typedef _MatrixType MatrixType;
00058 enum {
00059 RowsAtCompileTime = MatrixType::RowsAtCompileTime,
00060 ColsAtCompileTime = MatrixType::ColsAtCompileTime,
00061 Options = MatrixType::Options,
00062 MaxRowsAtCompileTime = MatrixType::MaxRowsAtCompileTime,
00063 MaxColsAtCompileTime = MatrixType::MaxColsAtCompileTime
00064 };
00065 typedef typename MatrixType::Scalar Scalar;
00066 typedef std::complex<typename NumTraits<Scalar>::Real> ComplexScalar;
00067 typedef typename MatrixType::Index Index;
00068
00069 typedef Matrix<ComplexScalar, ColsAtCompileTime, 1, Options & ~RowMajor, MaxColsAtCompileTime, 1> EigenvalueType;
00070 typedef Matrix<Scalar, ColsAtCompileTime, 1, Options & ~RowMajor, MaxColsAtCompileTime, 1> ColumnVectorType;
00071
00083 RealSchur(Index size = RowsAtCompileTime==Dynamic ? 1 : RowsAtCompileTime)
00084 : m_matT(size, size),
00085 m_matU(size, size),
00086 m_workspaceVector(size),
00087 m_hess(size),
00088 m_isInitialized(false),
00089 m_matUisUptodate(false),
00090 m_maxIters(-1)
00091 { }
00092
00103 RealSchur(const MatrixType& matrix, bool computeU = true)
00104 : m_matT(matrix.rows(),matrix.cols()),
00105 m_matU(matrix.rows(),matrix.cols()),
00106 m_workspaceVector(matrix.rows()),
00107 m_hess(matrix.rows()),
00108 m_isInitialized(false),
00109 m_matUisUptodate(false),
00110 m_maxIters(-1)
00111 {
00112 compute(matrix, computeU);
00113 }
00114
00126 const MatrixType& matrixU() const
00127 {
00128 eigen_assert(m_isInitialized && "RealSchur is not initialized.");
00129 eigen_assert(m_matUisUptodate && "The matrix U has not been computed during the RealSchur decomposition.");
00130 return m_matU;
00131 }
00132
00143 const MatrixType& matrixT() const
00144 {
00145 eigen_assert(m_isInitialized && "RealSchur is not initialized.");
00146 return m_matT;
00147 }
00148
00168 RealSchur& compute(const MatrixType& matrix, bool computeU = true);
00169
00187 template<typename HessMatrixType, typename OrthMatrixType>
00188 RealSchur& computeFromHessenberg(const HessMatrixType& matrixH, const OrthMatrixType& matrixQ, bool computeU);
00193 ComputationInfo info() const
00194 {
00195 eigen_assert(m_isInitialized && "RealSchur is not initialized.");
00196 return m_info;
00197 }
00198
00204 RealSchur& setMaxIterations(Index maxIters)
00205 {
00206 m_maxIters = maxIters;
00207 return *this;
00208 }
00209
00211 Index getMaxIterations()
00212 {
00213 return m_maxIters;
00214 }
00215
00221 static const int m_maxIterationsPerRow = 40;
00222
00223 private:
00224
00225 MatrixType m_matT;
00226 MatrixType m_matU;
00227 ColumnVectorType m_workspaceVector;
00228 HessenbergDecomposition<MatrixType> m_hess;
00229 ComputationInfo m_info;
00230 bool m_isInitialized;
00231 bool m_matUisUptodate;
00232 Index m_maxIters;
00233
00234 typedef Matrix<Scalar,3,1> Vector3s;
00235
00236 Scalar computeNormOfT();
00237 Index findSmallSubdiagEntry(Index iu, const Scalar& norm);
00238 void splitOffTwoRows(Index iu, bool computeU, const Scalar& exshift);
00239 void computeShift(Index iu, Index iter, Scalar& exshift, Vector3s& shiftInfo);
00240 void initFrancisQRStep(Index il, Index iu, const Vector3s& shiftInfo, Index& im, Vector3s& firstHouseholderVector);
00241 void performFrancisQRStep(Index il, Index im, Index iu, bool computeU, const Vector3s& firstHouseholderVector, Scalar* workspace);
00242 };
00243
00244
00245 template<typename MatrixType>
00246 RealSchur<MatrixType>& RealSchur<MatrixType>::compute(const MatrixType& matrix, bool computeU)
00247 {
00248 eigen_assert(matrix.cols() == matrix.rows());
00249 Index maxIters = m_maxIters;
00250 if (maxIters == -1)
00251 maxIters = m_maxIterationsPerRow * matrix.rows();
00252
00253
00254 m_hess.compute(matrix);
00255
00256
00257 computeFromHessenberg(m_hess.matrixH(), m_hess.matrixQ(), computeU);
00258
00259 return *this;
00260 }
00261 template<typename MatrixType>
00262 template<typename HessMatrixType, typename OrthMatrixType>
00263 RealSchur<MatrixType>& RealSchur<MatrixType>::computeFromHessenberg(const HessMatrixType& matrixH, const OrthMatrixType& matrixQ, bool computeU)
00264 {
00265 m_matT = matrixH;
00266 if(computeU)
00267 m_matU = matrixQ;
00268
00269 Index maxIters = m_maxIters;
00270 if (maxIters == -1)
00271 maxIters = m_maxIterationsPerRow * matrixH.rows();
00272 m_workspaceVector.resize(m_matT.cols());
00273 Scalar* workspace = &m_workspaceVector.coeffRef(0);
00274
00275
00276
00277
00278
00279 Index iu = m_matT.cols() - 1;
00280 Index iter = 0;
00281 Index totalIter = 0;
00282 Scalar exshift(0);
00283 Scalar norm = computeNormOfT();
00284
00285 if(norm!=0)
00286 {
00287 while (iu >= 0)
00288 {
00289 Index il = findSmallSubdiagEntry(iu, norm);
00290
00291
00292 if (il == iu)
00293 {
00294 m_matT.coeffRef(iu,iu) = m_matT.coeff(iu,iu) + exshift;
00295 if (iu > 0)
00296 m_matT.coeffRef(iu, iu-1) = Scalar(0);
00297 iu--;
00298 iter = 0;
00299 }
00300 else if (il == iu-1)
00301 {
00302 splitOffTwoRows(iu, computeU, exshift);
00303 iu -= 2;
00304 iter = 0;
00305 }
00306 else
00307 {
00308
00309 Vector3s firstHouseholderVector(0,0,0), shiftInfo;
00310 computeShift(iu, iter, exshift, shiftInfo);
00311 iter = iter + 1;
00312 totalIter = totalIter + 1;
00313 if (totalIter > maxIters) break;
00314 Index im;
00315 initFrancisQRStep(il, iu, shiftInfo, im, firstHouseholderVector);
00316 performFrancisQRStep(il, im, iu, computeU, firstHouseholderVector, workspace);
00317 }
00318 }
00319 }
00320 if(totalIter <= maxIters)
00321 m_info = Success;
00322 else
00323 m_info = NoConvergence;
00324
00325 m_isInitialized = true;
00326 m_matUisUptodate = computeU;
00327 return *this;
00328 }
00329
00331 template<typename MatrixType>
00332 inline typename MatrixType::Scalar RealSchur<MatrixType>::computeNormOfT()
00333 {
00334 const Index size = m_matT.cols();
00335
00336
00337
00338 Scalar norm(0);
00339 for (Index j = 0; j < size; ++j)
00340 norm += m_matT.col(j).segment(0, (std::min)(size,j+2)).cwiseAbs().sum();
00341 return norm;
00342 }
00343
00345 template<typename MatrixType>
00346 inline typename MatrixType::Index RealSchur<MatrixType>::findSmallSubdiagEntry(Index iu, const Scalar& norm)
00347 {
00348 using std::abs;
00349 Index res = iu;
00350 while (res > 0)
00351 {
00352 Scalar s = abs(m_matT.coeff(res-1,res-1)) + abs(m_matT.coeff(res,res));
00353 if (s == 0.0)
00354 s = norm;
00355 if (abs(m_matT.coeff(res,res-1)) < NumTraits<Scalar>::epsilon() * s)
00356 break;
00357 res--;
00358 }
00359 return res;
00360 }
00361
00363 template<typename MatrixType>
00364 inline void RealSchur<MatrixType>::splitOffTwoRows(Index iu, bool computeU, const Scalar& exshift)
00365 {
00366 using std::sqrt;
00367 using std::abs;
00368 const Index size = m_matT.cols();
00369
00370
00371
00372 Scalar p = Scalar(0.5) * (m_matT.coeff(iu-1,iu-1) - m_matT.coeff(iu,iu));
00373 Scalar q = p * p + m_matT.coeff(iu,iu-1) * m_matT.coeff(iu-1,iu);
00374 m_matT.coeffRef(iu,iu) += exshift;
00375 m_matT.coeffRef(iu-1,iu-1) += exshift;
00376
00377 if (q >= Scalar(0))
00378 {
00379 Scalar z = sqrt(abs(q));
00380 JacobiRotation<Scalar> rot;
00381 if (p >= Scalar(0))
00382 rot.makeGivens(p + z, m_matT.coeff(iu, iu-1));
00383 else
00384 rot.makeGivens(p - z, m_matT.coeff(iu, iu-1));
00385
00386 m_matT.rightCols(size-iu+1).applyOnTheLeft(iu-1, iu, rot.adjoint());
00387 m_matT.topRows(iu+1).applyOnTheRight(iu-1, iu, rot);
00388 m_matT.coeffRef(iu, iu-1) = Scalar(0);
00389 if (computeU)
00390 m_matU.applyOnTheRight(iu-1, iu, rot);
00391 }
00392
00393 if (iu > 1)
00394 m_matT.coeffRef(iu-1, iu-2) = Scalar(0);
00395 }
00396
00398 template<typename MatrixType>
00399 inline void RealSchur<MatrixType>::computeShift(Index iu, Index iter, Scalar& exshift, Vector3s& shiftInfo)
00400 {
00401 using std::sqrt;
00402 using std::abs;
00403 shiftInfo.coeffRef(0) = m_matT.coeff(iu,iu);
00404 shiftInfo.coeffRef(1) = m_matT.coeff(iu-1,iu-1);
00405 shiftInfo.coeffRef(2) = m_matT.coeff(iu,iu-1) * m_matT.coeff(iu-1,iu);
00406
00407
00408 if (iter == 10)
00409 {
00410 exshift += shiftInfo.coeff(0);
00411 for (Index i = 0; i <= iu; ++i)
00412 m_matT.coeffRef(i,i) -= shiftInfo.coeff(0);
00413 Scalar s = abs(m_matT.coeff(iu,iu-1)) + abs(m_matT.coeff(iu-1,iu-2));
00414 shiftInfo.coeffRef(0) = Scalar(0.75) * s;
00415 shiftInfo.coeffRef(1) = Scalar(0.75) * s;
00416 shiftInfo.coeffRef(2) = Scalar(-0.4375) * s * s;
00417 }
00418
00419
00420 if (iter == 30)
00421 {
00422 Scalar s = (shiftInfo.coeff(1) - shiftInfo.coeff(0)) / Scalar(2.0);
00423 s = s * s + shiftInfo.coeff(2);
00424 if (s > Scalar(0))
00425 {
00426 s = sqrt(s);
00427 if (shiftInfo.coeff(1) < shiftInfo.coeff(0))
00428 s = -s;
00429 s = s + (shiftInfo.coeff(1) - shiftInfo.coeff(0)) / Scalar(2.0);
00430 s = shiftInfo.coeff(0) - shiftInfo.coeff(2) / s;
00431 exshift += s;
00432 for (Index i = 0; i <= iu; ++i)
00433 m_matT.coeffRef(i,i) -= s;
00434 shiftInfo.setConstant(Scalar(0.964));
00435 }
00436 }
00437 }
00438
00440 template<typename MatrixType>
00441 inline void RealSchur<MatrixType>::initFrancisQRStep(Index il, Index iu, const Vector3s& shiftInfo, Index& im, Vector3s& firstHouseholderVector)
00442 {
00443 using std::abs;
00444 Vector3s& v = firstHouseholderVector;
00445
00446 for (im = iu-2; im >= il; --im)
00447 {
00448 const Scalar Tmm = m_matT.coeff(im,im);
00449 const Scalar r = shiftInfo.coeff(0) - Tmm;
00450 const Scalar s = shiftInfo.coeff(1) - Tmm;
00451 v.coeffRef(0) = (r * s - shiftInfo.coeff(2)) / m_matT.coeff(im+1,im) + m_matT.coeff(im,im+1);
00452 v.coeffRef(1) = m_matT.coeff(im+1,im+1) - Tmm - r - s;
00453 v.coeffRef(2) = m_matT.coeff(im+2,im+1);
00454 if (im == il) {
00455 break;
00456 }
00457 const Scalar lhs = m_matT.coeff(im,im-1) * (abs(v.coeff(1)) + abs(v.coeff(2)));
00458 const Scalar rhs = v.coeff(0) * (abs(m_matT.coeff(im-1,im-1)) + abs(Tmm) + abs(m_matT.coeff(im+1,im+1)));
00459 if (abs(lhs) < NumTraits<Scalar>::epsilon() * rhs)
00460 {
00461 break;
00462 }
00463 }
00464 }
00465
00467 template<typename MatrixType>
00468 inline void RealSchur<MatrixType>::performFrancisQRStep(Index il, Index im, Index iu, bool computeU, const Vector3s& firstHouseholderVector, Scalar* workspace)
00469 {
00470 eigen_assert(im >= il);
00471 eigen_assert(im <= iu-2);
00472
00473 const Index size = m_matT.cols();
00474
00475 for (Index k = im; k <= iu-2; ++k)
00476 {
00477 bool firstIteration = (k == im);
00478
00479 Vector3s v;
00480 if (firstIteration)
00481 v = firstHouseholderVector;
00482 else
00483 v = m_matT.template block<3,1>(k,k-1);
00484
00485 Scalar tau, beta;
00486 Matrix<Scalar, 2, 1> ess;
00487 v.makeHouseholder(ess, tau, beta);
00488
00489 if (beta != Scalar(0))
00490 {
00491 if (firstIteration && k > il)
00492 m_matT.coeffRef(k,k-1) = -m_matT.coeff(k,k-1);
00493 else if (!firstIteration)
00494 m_matT.coeffRef(k,k-1) = beta;
00495
00496
00497 m_matT.block(k, k, 3, size-k).applyHouseholderOnTheLeft(ess, tau, workspace);
00498 m_matT.block(0, k, (std::min)(iu,k+3) + 1, 3).applyHouseholderOnTheRight(ess, tau, workspace);
00499 if (computeU)
00500 m_matU.block(0, k, size, 3).applyHouseholderOnTheRight(ess, tau, workspace);
00501 }
00502 }
00503
00504 Matrix<Scalar, 2, 1> v = m_matT.template block<2,1>(iu-1, iu-2);
00505 Scalar tau, beta;
00506 Matrix<Scalar, 1, 1> ess;
00507 v.makeHouseholder(ess, tau, beta);
00508
00509 if (beta != Scalar(0))
00510 {
00511 m_matT.coeffRef(iu-1, iu-2) = beta;
00512 m_matT.block(iu-1, iu-1, 2, size-iu+1).applyHouseholderOnTheLeft(ess, tau, workspace);
00513 m_matT.block(0, iu-1, iu+1, 2).applyHouseholderOnTheRight(ess, tau, workspace);
00514 if (computeU)
00515 m_matU.block(0, iu-1, size, 2).applyHouseholderOnTheRight(ess, tau, workspace);
00516 }
00517
00518
00519 for (Index i = im+2; i <= iu; ++i)
00520 {
00521 m_matT.coeffRef(i,i-2) = Scalar(0);
00522 if (i > im+2)
00523 m_matT.coeffRef(i,i-3) = Scalar(0);
00524 }
00525 }
00526
00527 }
00528
00529 #endif // EIGEN_REAL_SCHUR_H