×

Solving stable Sylvester equations via rational iterative schemes. (English) Zbl 1098.65041

Summary: We investigate the numerical solution of stable Sylvester equations via iterative schemes proposed for computing the sign function of a matrix. In particular, we discuss how the rational iterations for the matrix sign function can efficiently be adapted to the special structure implied by the Sylvester equation. For Sylvester equations with factored constant term as those arising in model reduction or image restoration, we derive an algorithm that computes the solution in factored form directly. We also suggest convergence criteria for the resulting iterations and compare the accuracy and performance of the resulting methods with existing Sylvester solvers. The algorithms proposed here are easy to parallelize. We report on the parallelization of those algorithms and demonstrate their high efficiency and scalability using experimental results obtained on a cluster of Intel Pentium Xeon processors.

MSC:

65F30 Other matrix algorithms (MSC2010)
15A24 Matrix equations and identities
65Y05 Parallel numerical computation
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Abels J., and Benner, P. (1999). CAREX–a collection of benchmark examples for continuous-time algebraic Riccati equations (version 2.0). SLICOT Working Note 1999-14. Available from http://www.win.tue.nl/niconet/NIC2/reports.html
[2] Aldhaheri R. (1991). Model order reduction via real Schur-form decomposition. Int. J. Control 53(3):709–716 · Zbl 0725.93015 · doi:10.1080/00207179108953642
[3] Anderson E., Bai Z., Bischof C., Demmel J., Dongarra J., Du Croz J., Greenbaum A., Hammarling S., McKenney A., Sorensen D. (1999). LAPACK Users’ Guide 3rd ed. SIAM, Philadelphia, PA · Zbl 0934.65030
[4] Antoulas A. (2005). pproximation of Large-Scale Dynamical Systems. SIAM Publications, Philadelphia, PA · Zbl 1112.93002
[5] Bai, Z., and Demmel, J. (1993). Design of a parallel nonsymmetric eigenroutine toolbox, Part I. In R. S. et al., (eds.), Proceedings of the Sixth SIAM Conference on Parallel Processing for Scientific Computing, SIAM, Philadelphia, PA, 391–398, See also: Tech. Report CSD-92-718, Computer Science Division, University of California, Berkeley, CA.
[6] Bartels R., Stewart G. (1972). Solution of the matrix equation AX+XB=C: Algorithm 432. Comm. ACM 15:820–826 · Zbl 1372.65121 · doi:10.1145/361573.361582
[7] Beavers A.N., Denman E.D. (1975). A new solution method for the Lyapunov matrix equations. SIAM J. Appl. Math. 29:416–42 · Zbl 0317.15011 · doi:10.1137/0129035
[8] Benner P. (2004). Factorized solution of Sylvester equations with applications in control. In Proc. Intl. Symp. Math. Theory Networks and Syst. MTNS 2004, http://www.mtns2004.be
[9] Benner P., Claver v., Quintana-Ortí E. (1999). Parallel distributed solvers for large stable generalized Lyapunov equations. Parallel Processing Lett. 9(1):147–158 · doi:10.1142/S0129626499000165
[10] Benner P., Quintana-Ortí E. (1999). Solving stable generalized Lyapunov equations with the matrix sign function. Numer. Algorithms 20(1):75–100 · Zbl 0940.65035 · doi:10.1023/A:1019191431273
[11] Benner P., Quintana-Ortí E., Quintana-Ortí G. (2002). Numerical solution of discrete stable linear matrix equations on multicomputers. Parallel Algorithms and Appl. 17(1):127–146 · Zbl 1009.65039
[12] Blackford L., Choi J., Cleary A., D’Azevedo E., Demmel J., Dhillon I., Dongarra J., Hammarling S., Henry G., Petitet A., Stanley K., Walker D., Whaley R. (1997). ScaLAPACK Users’ Guide. SIAM, Philadelphia, PA · Zbl 0886.65022
[13] Byers R. (1987). Solving the algebraic Riccati equation with the matrix sign function. Linear Algebra Appl. 85:267–279 · Zbl 0611.65027 · doi:10.1016/0024-3795(87)90222-9
[14] Calvetti D., Reichel L. (1996). Application of ADI iterative methods to the restoration of noisy images. SIAM J. Matrix Anal. Appl. 17:165–186 · Zbl 0849.65101 · doi:10.1137/S0895479894273687
[15] Chan T. (1987). Rank revealing QR factorizations. Linear Algebra Appl. 88/89:67–82 · Zbl 0624.65025 · doi:10.1016/0024-3795(87)90103-0
[16] Choi C., Laub A. (1990). Efficient matrix-valued algorithms for solving stiff Riccati differential equations. IEEE Trans. Automat. Control 35:770–776 · Zbl 0714.93011 · doi:10.1109/9.57015
[17] Datta B. (2003). Numerical Methods for Linear Control Systems Design and Analysis. Elsevier Press, Amsterdam
[18] Dieci L., Osborne M., Russell R. (1988). A. Riccati, transformation method for solving linear BVPs I: Theoretical aspects. SIAM J. Numer. Anal. 25(5):1055–1073 · Zbl 0664.65074 · doi:10.1137/0725061
[19] Dongarra J., Croz J.D., Duff I., Hammarling S. (1990). A set of level 3 Basic linear algebra subprograms. ACM Trans. Math. Soft. 16:1–17 · Zbl 0900.65115 · doi:10.1145/77626.79170
[20] Dongarra J., Duff I.S., Sorensen D.C., van der Vorst H.A. (1991). Solving Linear Systems on Vector and Shared Memory computers. SIAM, Philadelphia, PA · Zbl 0770.65009
[21] Enright W. (1978). Improving the efficiency of matrix operations in the numerical solution of stiff ordinary differential equations. ACM Trans. Math. Soft. 4:127–136 · Zbl 0382.65029 · doi:10.1145/355780.355784
[22] Epton M. (1980). Methods for the solution of AXDXC=E and its application in the numerical solution of implicit ordinary differential equations. BIT 20:341–345 · Zbl 0452.65015 · doi:10.1007/BF01932775
[23] Fernando K., Nicholson H. (1984). On a fundamental property of the cross-Gramian matrix. IEEE Trans. Circuits Syst. CAS- 31(5): 504–505 · Zbl 0538.93008 · doi:10.1109/TCS.1984.1085524
[24] Gajić Z., Qureshi M. (1995). Lyapunov Matrix Equation in System Stability and Control. Math. in Science and Engineering. Academic Press, San Diego, CA
[25] Gardiner J., Laub A. (1991). Parallel algorithms for algebraic Riccati equations. Int. J. Control 54(6):1317–1333 · Zbl 0763.93036 · doi:10.1080/00207179108934214
[26] Gardiner J., Laub A., Amato J., Moler C. (1992). Solution of the Sylvester matrix equation AXB+CXD=E. ACM Trans. Math. Soft 18:223–231 · Zbl 0893.65026 · doi:10.1145/146847.146929
[27] Golub G., Van Loan C. (1996). Matrix Computations 3rd ed. Johns Hopkins University Press, Baltimore · Zbl 0865.65009
[28] Golub G.H., Nash S., Van Loan C.F. (1979). A Hessenberg–Schur method for the problem AX+XB=C. IEEE Trans. Automat. Control. AC- 24:909–913 · Zbl 0421.65022 · doi:10.1109/TAC.1979.1102170
[29] Goto K., van de Geijn R. (2002). On reducing TLB misses in matrix multiplication. FLAME Working Note 9, Department of Computer Sciences, The University of Texas at Austin, http://www.cs.utexas.edu/users/flame
[30] Grasedyck L. (2004). Existence of a low rank or H-matrix approximant to the solution of a Sylvester equation. Numer. Lin. Alg. Appl. 11:371–389 · Zbl 1164.65381 · doi:10.1002/nla.366
[31] Grasedyck L., Hackbusch W. (2004). A multigrid method to solve large scale Sylvester equations. Preprint 48, Max-Planck Institut für Mathematik in den Naturwissenschaften, Leipzig Germany · Zbl 1154.65024
[32] Gropp W., Lusk E., Skjellum A. (1994). Using MPI: Portable Parallel Programming with the Message-Passing Interface. MIT Press, Cambridge, MA · Zbl 0875.68206
[33] Halley E. (1694). Methodus nova, accurata & facilis inveniendi radices aequationum quarumcumque generaliter, sine praevia reductione. Philos. Trans. R. Soc. Lond 18:136–148 · doi:10.1098/rstl.1694.0029
[34] Henry G., van de Geijn R. (1997). Parallelizing the QR algorithm for the unsymmetric algebraic eigenvalue problem: myths and reality. SIAM J. Sci. Comput. 17:870–883 · Zbl 0856.65027 · doi:10.1137/0917056
[35] Higham N. (1986). Computing the polar decomposition–with applications. SIAM J. Sci. Statist. Comput. 7:1160–1174 · Zbl 0607.65014 · doi:10.1137/0907079
[36] Hoskins W., Meek D., Walton D. (1977). The numerical solution of the matrix equation X A+A Y=F. BIT 17:184–190 · Zbl 0358.65025 · doi:10.1007/BF01932289
[37] Hu D., Reichel L. (1992). Krylov-subspace methods for the Sylvester equation. Linear Algebra Appl. 172:283–313 · Zbl 0777.65028 · doi:10.1016/0024-3795(92)90031-5
[38] Kågström B., Poromaa P. (1996). Lapack-style algorithms and software for solving the generalized Sylvester equation and estimating the separation between regular matrix pairs. ACM Trans. Math. Soft 22(1):78–103 · Zbl 0884.65031 · doi:10.1145/225545.225552
[39] Kenney C., and Laub A. (1991). Rational iterative methods for the matrix sign function. SIAM J. Matrix Anal. Appl. 12: 273–291 · Zbl 0725.65048 · doi:10.1137/0612020
[40] Kenney C., Laub A. (1995). The matrix sign function. IEEE Trans Automat Control 40(8):1330–1348 · Zbl 0830.93022 · doi:10.1109/9.402226
[41] Quintana-Ortí E., Quintana-Ortí G., Sun X., van de Geijn R. (2001). A note on parallel matrix inversion. SIAM J. Sci. Comput. 22:1762–1771 · Zbl 0991.65029 · doi:10.1137/S1064827598345679
[42] Roberts, J. (1980). Linear model reduction and solution of the algebraic Riccati equation by use of the sign function. Int. J. Control, 32, 677–687, (Reprint of Technical Report No. TR-13, CUED/B-Control, Cambridge University, Engineering Department, 1971). · Zbl 0463.93050
[43] Schulz G. (1993). Iterative berechnung der reziproken matrix. Z. Angew. Math. Mech. 13:57–59 In German · JFM 59.0535.04 · doi:10.1002/zamm.19330130111
[44] Sima V. (1996). Algorithms for Linear-Quadratic Optimization. Pure and Applied Mathematics vol 200. Marcel Dekker Inc., New York NY · Zbl 0863.65038
[45] Slowik, M., Benner, P., and Sima, V. (2004). Evaluation of the linear matrix equation solvers in SLICOT. SLICOT Working Note 2004–1 Available from http://www.win.tue.nl/niconet/NIC2/reports.html · Zbl 1133.93317
[46] van de Geijn R. (1997). Using PLAPACK: Parallel Linear Algebra Package. MIT Press, Cambridge MA
[47] Wachspress E. (1988). Iterative solution of the Lyapunov matrix equation. Appl. Math. Lett. 107:87–90 · Zbl 0631.65037 · doi:10.1016/0893-9659(88)90183-8
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.