Abstract
A Schur-type algorithm is presented for the simultaneous triangular factorization of a given (non-degenerate) structured matrix and its inverse. The algorithm takes the displacement generator of a Hermitian, strongly regular matrixR as an input, and computes the displacement generator of the inverse matrixR −1 as an output. From these generators we can directly deduce theLD −1 L * (lower-diagonal-upper) decomposition ofR, and theUD −1 U * (upper-diagonallower) decomposition ofR −1. The computational complexity of the algorithm isO(rn 2) operations, wheren andr denote the size and the displacement rank ofR, respectively. Moreover, this method is especially suited for parallel (systolic array) implementations: usingn processors the algorithm can be carried out inO(n) steps.
Similar content being viewed by others
References
T. Boros, Studies in Displacement Structure Theory. PhD thesis, Stanford University, Stanford, CA, June 1996.
T. Boros, A.H. Sayed, T. Kailath, and H. Lev-Ari,A generalized Schurtype algorithm for the joint factorization of a structured matrix and its inverse, Part I: Non-degenerate case, In preparation.
T. Boros, A. H. Sayed, T. Kailath, and H. Lev-Ari,A generalized Schurtype algorithm for the joint factorization of a structured matrix and its inverse, Part II: Degenerate case, In preparation.
J. Chun, Fast Array Algorithms for Structured Matrices. PhD thesis, Stanford University, Stanford, CA, 1989.
J. Chun andT. Kailath,Displacement structure for Hankel, Vandermonde and related (derived) matrices, Lin. Alg. and Its Appl.151, (1991) 199–227.
J. M. Delosme, Algorithms and Implementations for Linear Least-Squares Estimation. PhD thesis, Stanford University, Stanford, CA, 1982.
P. Delsarte, Y. Genin, andY. Kamp,A polynomial approach to the generalized Levinson algorithm. IEEE Trans. Inform. Theory29(2, (March 1983) 268–278.
B. Friedlander, T. Kailath, M. Morf, andL. Ljung,Extended Levinson and Chandrasekhar equations for general discrete-time linear estimation problems. IEEE Trans. Automatic Control23(4), (August 1978) 653–659.
Y. Genin, P. Van Dooren, T. Kailath, J. Delosme, andM. Morf,On Σ-lossless transfer functions and related questions. Lin. Alg. and Its Appl.50, (1983) 251–275.
G. H. Golub and C. F. Van Loan, Matrix Computation. The Johns Hopkins University Press, 2-nd edition, 1989,
T. Kailath,Signal processing in the VLSI era. In S. Y. Kung, H. Whitehouse, and T. Kailath, editors, VLSI and Modern Sign. Proc., pp. 5–24. Prentice Hall, 1985.
T. Kailath,Signal processing applications of some moment problems. In Proc. of Symp. in Appl. Math. vol.37, pp. 71–109. American Math. Soc., 1987.
T. Kailath andJ. Chun,Generalized displacement structure for block-Toeplitz, Toeplitz-block, and Toeplitz-derived matrices. SIAM J. Matrix Anal. Appl.15(1), (January 1994) 114–128.
S. Y. Kung andY. H. Hu,A highly concurrent algorithm and pipelined architecture for solving Toeplitz systems. IEEE Trans. on Acoust., Speech, and Sign. Proc.31, (1983) 66–76.
T. Kailath, S. Y. Kung, andM. Morf,Displacement ranks of a matrix. Bulletin of the American Mathematical Society1(5), (September 1979) 769–773.
T. Kailath andA. H. Sayed,Displacement structure: Theory and applications. SIAM Review37(3), (1995) 297–386.
H. Lev-Ari, Nonstationary Lattice-Filter Modeling. PhD thesis, Stanford University, Stanford, CA, December 1983.
H. Lev-Ari andT. Kailath,Lattice filter parametrization and modeling of nonstationary processes. IEEE Trans. Inform. Theory30, (1984) 2–16.
H. Lev-Ari andT. Kailath,Triangular factorization of structured Hermitian matrices. Operator Theory: Advances and Applications18, (1986) 301–324.
H. Lev-Ari andT. Kailath,State-space approach to factorization of lossless transfer functions and structured matrices. Lin. Alg. and Its Appl.162–164, (1992) 273–295.
A.H. Sayed, Displacement Structure in Signal Processing and Mathematics. PhD thesis, Stanford University, Stanford, CA, August 1992.
A. H. Sayed, T. Constantinescu, and T. Kailath,Square-root algorithms for structured matrices, interpolation, and completion problems. In Lin. Alg. for Sign. Proc., vol.69 of IMA Volumes in Math. and its Appl., pp. 153–184. Springer, 1995.
A. H. Sayed, H. Lev-Ari, andT. Kailath,Time-varying displacement structure and triangular arrays. IEEE Trans. Signal Proc.42, (May 1994) 1052–1062.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Boros, T., Sayed, A.H., Lev-Ari, H. et al. A generalized Schur-type algorithm for the joint factorization of a structured matrix and its inverse. Calcolo 33, 131–145 (1996). https://doi.org/10.1007/BF02575713
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02575713