The Matrix Sign Function Method and the Computation of Invariant Subspaces

A perturbation analysis shows that if a numerically stable procedure is used to compute the matrix sign function, then it is competitive with conventional methods for computing invariant subspaces. Stability analysis of the Newton iteration improves an earlier result of Byers and confirms that ill-conditioned iterates may cause numerical instability. Numerical examples demonstrate the theoretical results.

[1]  B. Nagy.  Perturbations des transformations autoadjointes dans l'espace de Hilbert , 1946 .

[2]  Tosio Kato,et al.  On the Convergence of the Perturbation Method, II. 1 , 1950 .

[3]  Tosio Kato,et al.  On the convergence of the perturbation method , 1951 .

[4]  M. Rosenblum,et al.  On the operator equation $BX-XA=Q$ , 1956 .

[5]  Tosio Kato.  Perturbation theory for linear operators , 1966 .

[6]  Peter Lancaster,et al.  The theory of matrices , 1969 .

[7]  William Kahan,et al.  A Survey of Error Analysis , 1971, IFIP Congress.

[8]  G. Stewart.  Error and Perturbation Bounds for Subspaces Associated with Certain Eigenvalue Problems , 1973 .

[9]  G. Stewart.  Introduction to matrix computations , 1973 .

[10]  E. Denman,et al.  A computational method for eigenvalues and eigenvectors of a matrix with real eigenvalues , 1973 .

[11]  E. Denman,et al.  The matrix sign function and computations in systems , 1976 .

[12]  L. Balzer.  Accelerated convergence of the matrix sign function method of solving Lyapunov, Riccati and other matrix equations , 1980 .

[13]  Webb Miller,et al.  Software for Roundoff Analysis of Matrix Algorithms , 1980 .

[14]  J. D. Roberts,et al.  Linear model reduction and solution of the algebraic Riccati equation by use of the sign function , 1980 .

[15]  J. Leyva-Ramos,et al.  Spectral decomposition of a matrix using the generalized sign matrix , 1981 .

[16]  Judith Gardiner,et al.  A generalization of the matrix sign function solution for algebraic Riccati equations , 1985, 1985 24th IEEE Conference on Decision and Control.

[17]  R. Byers.  Numerical Stability and Instability in Matrix Sign Function Based Algorithms , 1986 .

[18]  N. Higham.  Computing the polar decomposition with applications , 1986 .

[19]  N. Higham.  Newton's method for the matrix square root , 1986 .

[20]  R. Byers.  Solving the algebraic Riccati equation with the matrix sign function , 1987 .

[21]  R. Byers.  A Bisection Method for Measuring the Distance of a Stable Matrix to the Unstable Matrices , 1988 .

[22]  Paul Van Dooren,et al.  A systolic algorithm for riccati and lyapunov equations , 1989, Math. Control. Signals Syst..

[23]  Alan J. Laub,et al.  On Scaling Newton's Method for Polar Decomposition and the Matrix Sign Function , 1990, 1990 American Control Conference.

[24]  Alan J. Laub,et al.  A Parallel Algorithm for the Matrix Sign Function , 1990, Int. J. High Speed Comput..

[25]  A. Laub,et al.  Parallel algorithms for algebraic Riccati equations , 1991 .

[26]  V. N. Bogaevski,et al.  Matrix Perturbation Theory , 1991 .

[27]  Charles Kenney,et al.  Polar Decomposition and Matrix Sign Function Condition Estimates , 1991, SIAM J. Sci. Comput..

[28]  A. Laub,et al.  Rational iterative methods for the matrix sign function , 1991 .

[29]  A. Laub,et al.  Matrix-sign algorithms for Riccati equations , 1992 .

[30]  James Demmel,et al.  Design of a Parallel Nonsymmetric Eigenroutine Toolbox, Part I , 1993, PPSC.

[31]  A. Laub,et al.  The matrix sign function , 1995, IEEE Trans. Autom. Control..

[32]  J. Demmel,et al.  An inverse free parallel spectral divide and conquer algorithm for nonsymmetric eigenproblems , 1997 .