1CTGEXC(1)                LAPACK routine (version 3.1)                CTGEXC(1)
2
3
4

NAME

6       CTGEXC  -  the generalized Schur decomposition of a complex matrix pair
7       (A,B), using an unitary equivalence transformation (A, B) := Q * (A, B)
8       * Z', so that the diagonal block of (A, B) with row index IFST is moved
9       to row ILST
10

SYNOPSIS

12       SUBROUTINE CTGEXC( WANTQ, WANTZ, N, A, LDA, B, LDB,  Q,  LDQ,  Z,  LDZ,
13                          IFST, ILST, INFO )
14
15           LOGICAL        WANTQ, WANTZ
16
17           INTEGER        IFST, ILST, INFO, LDA, LDB, LDQ, LDZ, N
18
19           COMPLEX        A( LDA, * ), B( LDB, * ), Q( LDQ, * ), Z( LDZ, * )
20

PURPOSE

22       CTGEXC reorders the generalized Schur decomposition of a complex matrix
23       pair (A,B), using an unitary equivalence transformation (A, B) :=  Q  *
24       (A,  B)  * Z', so that the diagonal block of (A, B) with row index IFST
25       is moved to row ILST.
26
27       (A, B) must be in generalized Schur canonical form, that is,  A  and  B
28       are both upper triangular.
29
30       Optionally,  the  matrices  Q  and  Z  of generalized Schur vectors are
31       updated.
32
33              Q(in) * A(in) * Z(in)' = Q(out) * A(out) * Z(out)'
34              Q(in) * B(in) * Z(in)' = Q(out) * B(out) * Z(out)'
35
36

ARGUMENTS

38       WANTQ   (input) LOGICAL
39
40       WANTZ   (input) LOGICAL
41
42       N       (input) INTEGER
43               The order of the matrices A and B. N >= 0.
44
45       A       (input/output) COMPLEX array, dimension (LDA,N)
46               On entry, the upper triangular matrix A in the pair (A, B).  On
47               exit, the updated matrix A.
48
49       LDA     (input)  INTEGER
50               The leading dimension of the array A. LDA >= max(1,N).
51
52       B       (input/output) COMPLEX array, dimension (LDB,N)
53               On entry, the upper triangular matrix B in the pair (A, B).  On
54               exit, the updated matrix B.
55
56       LDB     (input)  INTEGER
57               The leading dimension of the array B. LDB >= max(1,N).
58
59       Q       (input/output) COMPLEX array, dimension (LDZ,N)
60               On entry, if WANTQ = .TRUE., the unitary matrix  Q.   On  exit,
61               the updated matrix Q.  If WANTQ = .FALSE., Q is not referenced.
62
63       LDQ     (input) INTEGER
64               The  leading  dimension  of  the  array Q. LDQ >= 1; If WANTQ =
65               .TRUE., LDQ >= N.
66
67       Z       (input/output) COMPLEX array, dimension (LDZ,N)
68               On entry, if WANTZ = .TRUE., the unitary matrix  Z.   On  exit,
69               the updated matrix Z.  If WANTZ = .FALSE., Z is not referenced.
70
71       LDZ     (input) INTEGER
72               The  leading  dimension  of  the  array Z. LDZ >= 1; If WANTZ =
73               .TRUE., LDZ >= N.
74
75       IFST    (input) INTEGER
76               ILST    (input/output) INTEGER Specify the  reordering  of  the
77               diagonal  blocks  of  (A, B).  The block with row index IFST is
78               moved to row ILST, by a sequence of swapping  between  adjacent
79               blocks.
80
81       INFO    (output) INTEGER
82               =0:  Successful exit.
83               <0:  if INFO = -i, the i-th argument had an illegal value.
84               =1:   The  transformed matrix pair (A, B) would be too far from
85               generalized Schur form; the problem is ill- conditioned. (A, B)
86               may have been partially reordered, and ILST points to the first
87               row of the current position of the block being moved.
88

FURTHER DETAILS

90       Based on contributions by
91          Bo Kagstrom and Peter Poromaa, Department of Computing Science,
92          Umea University, S-901 87 Umea, Sweden.
93
94       [1] B. Kagstrom; A Direct Method for Reordering Eigenvalues in the
95           Generalized Real Schur Form of a Regular Matrix Pair (A, B), in
96           M.S. Moonen et al (eds), Linear Algebra for Large Scale and
97           Real-Time Applications, Kluwer Academic Publ. 1993, pp 195-218.
98
99       [2] B. Kagstrom and P. Poromaa; Computing Eigenspaces with Specified
100           Eigenvalues of a Regular Matrix Pair (A, B) and Condition
101           Estimation: Theory, Algorithms and Software, Report
102           UMINF - 94.04, Department of Computing Science, Umea University,
103           S-901 87 Umea, Sweden, 1994. Also as LAPACK Working Note 87.
104           To appear in Numerical Algorithms, 1996.
105
106       [3] B. Kagstrom and P. Poromaa, LAPACK-Style Algorithms and Software
107           for Solving the Generalized Sylvester Equation and Estimating the
108           Separation between Regular Matrix Pairs, Report UMINF - 93.23,
109           Department of Computing Science, Umea University, S-901 87 Umea,
110           Sweden, December 1993, Revised April 1994, Also as LAPACK working
111           Note 75. To appear in ACM Trans. on Math. Software, Vol 22, No 1,
112           1996.
113
114
115
116
117 LAPACK routine (version 3.1)    November 2006                       CTGEXC(1)
Impressum