1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
|
<HTML>
<HEAD><TITLE>MB03IZ - SLICOT Library Routine Documentation</TITLE>
</HEAD>
<BODY>
<H2><A Name="MB03IZ">MB03IZ</A></H2>
<H3>
Moving eigenvalues with negative real parts of a complex skew-Hamiltonian/Hamiltonian pencil in structured Schur form to the leading subpencil (factored version)
</H3>
<A HREF ="#Specification"><B>[Specification]</B></A>
<A HREF ="#Arguments"><B>[Arguments]</B></A>
<A HREF ="#Method"><B>[Method]</B></A>
<A HREF ="#References"><B>[References]</B></A>
<A HREF ="#Comments"><B>[Comments]</B></A>
<A HREF ="#Example"><B>[Example]</B></A>
<P>
<B><FONT SIZE="+1">Purpose</FONT></B>
<PRE>
To move the eigenvalues with strictly negative real parts of an
N-by-N complex skew-Hamiltonian/Hamiltonian pencil aS - bH in
structured Schur form, with
( 0 I )
S = J Z' J' Z, where J = ( ),
( -I 0 )
to the leading principal subpencil, while keeping the triangular
form. On entry, we have
( A D ) ( B F )
Z = ( ), H = ( ),
( 0 C ) ( 0 -B' )
where A and B are upper triangular and C is lower triangular.
Z and H are transformed by a unitary symplectic matrix U and a
unitary matrix Q such that
( Aout Dout )
Zout = U' Z Q = ( ), and
( 0 Cout )
(1)
( Bout Fout )
Hout = J Q' J' H Q = ( ),
( 0 -Bout' )
where Aout, Bout and Cout remain in triangular form. The notation
M' denotes the conjugate transpose of the matrix M.
Optionally, if COMPQ = 'I' or COMPQ = 'U', the unitary matrix Q
that fulfills (1) is computed.
Optionally, if COMPU = 'I' or COMPU = 'U', the unitary symplectic
matrix
( U1 U2 )
U = ( )
( -U2 U1 )
that fulfills (1) is computed.
</PRE>
<A name="Specification"><B><FONT SIZE="+1">Specification</FONT></B></A>
<PRE>
SUBROUTINE MB03IZ( COMPQ, COMPU, N, A, LDA, C, LDC, D, LDD, B,
$ LDB, F, LDF, Q, LDQ, U1, LDU1, U2, LDU2, NEIG,
$ TOL, INFO )
C .. Scalar Arguments ..
CHARACTER COMPQ, COMPU
INTEGER INFO, LDA, LDB, LDC, LDD, LDF, LDQ, LDU1, LDU2,
$ N, NEIG
DOUBLE PRECISION TOL
C .. Array Arguments ..
COMPLEX*16 A( LDA, * ), B( LDB, * ), C( LDC, * ),
$ D( LDD, * ), F( LDF, * ), Q( LDQ, * ),
$ U1( LDU1, * ), U2( LDU2, * )
</PRE>
<A name="Arguments"><B><FONT SIZE="+1">Arguments</FONT></B></A>
<P>
<B>Mode Parameters</B>
<PRE>
COMPQ CHARACTER*1
Specifies whether or not the unitary transformations
should be accumulated in the array Q, as follows:
= 'N': Q is not computed;
= 'I': the array Q is initialized internally to the unit
matrix, and the unitary matrix Q is returned;
= 'U': the array Q contains a unitary matrix Q0 on
entry, and the matrix Q0*Q is returned, where Q
is the product of the unitary transformations
that are applied to the pencil aS - bH to reorder
the eigenvalues.
COMPU CHARACTER*1
Specifies whether or not the unitary symplectic
transformations should be accumulated in the arrays U1 and
U2, as follows:
= 'N': U1 and U2 are not computed;
= 'I': the arrays U1 and U2 are initialized internally,
and the submatrices U1 and U2 defining the
unitary symplectic matrix U are returned;
= 'U': the arrays U1 and U2 contain the corresponding
submatrices of a unitary symplectic matrix U0
on entry, and the updated submatrices U1 and U2
of the matrix product U0*U are returned, where U
is the product of the unitary symplectic
transformations that are applied to the pencil
aS - bH to reorder the eigenvalues.
</PRE>
<B>Input/Output Parameters</B>
<PRE>
N (input) INTEGER
The order of the pencil aS - bH. N >= 0, even.
A (input/output) COMPLEX*16 array, dimension (LDA, N/2)
On entry, the leading N/2-by-N/2 part of this array must
contain the upper triangular matrix A.
On exit, the leading N/2-by-N/2 part of this array
contains the transformed matrix Aout.
The strictly lower triangular part of this array is not
referenced.
LDA INTEGER
The leading dimension of the array A. LDA >= MAX(1, N/2).
C (input/output) COMPLEX*16 array, dimension (LDC, N/2)
On entry, the leading N/2-by-N/2 part of this array must
contain the lower triangular matrix C.
On exit, the leading N/2-by-N/2 part of this array
contains the transformed matrix Cout.
The strictly upper triangular part of this array is not
referenced.
LDC INTEGER
The leading dimension of the array C. LDC >= MAX(1, N/2).
D (input/output) COMPLEX*16 array, dimension (LDD, N/2)
On entry, the leading N/2-by-N/2 part of this array must
contain the matrix D.
On exit, the leading N/2-by-N/2 part of this array
contains the transformed matrix Dout.
LDD INTEGER
The leading dimension of the array D. LDD >= MAX(1, N/2).
B (input/output) COMPLEX*16 array, dimension (LDB, N/2)
On entry, the leading N/2-by-N/2 part of this array must
contain the upper triangular matrix B.
On exit, the leading N/2-by-N/2 part of this array
contains the transformed matrix Bout.
The strictly lower triangular part of this array is not
referenced.
LDB INTEGER
The leading dimension of the array B. LDB >= MAX(1, N/2).
F (input/output) COMPLEX*16 array, dimension (LDF, N/2)
On entry, the leading N/2-by-N/2 part of this array must
contain the upper triangular part of the Hermitian matrix
F.
On exit, the leading N/2-by-N/2 part of this array
contains the transformed matrix Fout.
The strictly lower triangular part of this array is not
referenced.
LDF INTEGER
The leading dimension of the array F. LDF >= MAX(1, N/2).
Q (input/output) COMPLEX*16 array, dimension (LDQ, N)
On entry, if COMPQ = 'U', then the leading N-by-N part of
this array must contain a given matrix Q0, and on exit,
the leading N-by-N part of this array contains the product
of the input matrix Q0 and the transformation matrix Q
used to transform the matrices S and H.
On exit, if COMPQ = 'I', then the leading N-by-N part of
this array contains the unitary transformation matrix Q.
If COMPQ = 'N' this array is not referenced.
LDQ INTEGER
The leading dimension of the array Q.
LDQ >= 1, if COMPQ = 'N';
LDQ >= MAX(1, N), if COMPQ = 'I' or COMPQ = 'U'.
U1 (input/output) COMPLEX*16 array, dimension (LDU1, N/2)
On entry, if COMPU = 'U', then the leading N/2-by-N/2 part
of this array must contain the upper left block of a
given matrix U0, and on exit, the leading N/2-by-N/2 part
of this array contains the updated upper left block U1 of
the product of the input matrix U0 and the transformation
matrix U used to transform the matrices S and H.
On exit, if COMPU = 'I', then the leading N/2-by-N/2 part
of this array contains the upper left block U1 of the
unitary symplectic transformation matrix U.
If COMPU = 'N' this array is not referenced.
LDU1 INTEGER
The leading dimension of the array U1.
LDU1 >= 1, if COMPU = 'N';
LDU1 >= MAX(1, N/2), if COMPU = 'I' or COMPU = 'U'.
U2 (input/output) COMPLEX*16 array, dimension (LDU2, N/2)
On entry, if COMPU = 'U', then the leading N/2-by-N/2 part
of this array must contain the upper right block of a
given matrix U0, and on exit, the leading N/2-by-N/2 part
of this array contains the updated upper right block U2 of
the product of the input matrix U0 and the transformation
matrix U used to transform the matrices S and H.
On exit, if COMPU = 'I', then the leading N/2-by-N/2 part
of this array contains the upper right block U2 of the
unitary symplectic transformation matrix U.
If COMPU = 'N' this array is not referenced.
LDU2 INTEGER
The leading dimension of the array U2.
LDU2 >= 1, if COMPU = 'N';
LDU2 >= MAX(1, N/2), if COMPU = 'I' or COMPU = 'U'.
NEIG (output) INTEGER
The number of eigenvalues in aS - bH with strictly
negative real part.
</PRE>
<B>Tolerances</B>
<PRE>
TOL DOUBLE PRECISION
The tolerance used to decide the sign of the eigenvalues.
If the user sets TOL > 0, then the given value of TOL is
used. If the user sets TOL <= 0, then an implicitly
computed, default tolerance, defined by MIN(N,10)*EPS, is
used instead, where EPS is the machine precision (see
LAPACK Library routine DLAMCH). A larger value might be
needed for pencils with multiple eigenvalues.
</PRE>
<B>Error Indicator</B>
<PRE>
INFO INTEGER
= 0: succesful exit;
< 0: if INFO = -i, the i-th argument had an illegal value.
</PRE>
<A name="Method"><B><FONT SIZE="+1">Method</FONT></B></A>
<PRE>
The algorithm reorders the eigenvalues like the following scheme:
Step 1: Reorder the eigenvalues in the subpencil aC'*A - bB.
I. Reorder the eigenvalues with negative real parts to the
top.
II. Reorder the eigenvalues with positive real parts to the
bottom.
Step 2: Reorder the remaining eigenvalues with negative real
parts.
I. Exchange the eigenvalues between the last diagonal block
in aC'*A - bB and the last diagonal block in aS - bH.
II. Move the eigenvalues in the N/2-th place to the (MM+1)-th
place, where MM denotes the current number of eigenvalues
with negative real parts in aC'*A - bB.
The algorithm uses a sequence of unitary transformations as
described on page 38 in [1]. To achieve those transformations the
elementary SLICOT Library subroutines MB03CZ and MB03GZ are called
for the corresponding matrix structures.
</PRE>
<A name="References"><B><FONT SIZE="+1">References</FONT></B></A>
<PRE>
[1] Benner, P., Byers, R., Mehrmann, V. and Xu, H.
Numerical Computation of Deflating Subspaces of Embedded
Hamiltonian Pencils.
Tech. Rep. SFB393/99-15, Technical University Chemnitz,
Germany, June 1999.
</PRE>
<A name="Numerical Aspects"><B><FONT SIZE="+1">Numerical Aspects</FONT></B></A>
<PRE> 3
The algorithm is numerically backward stable and needs O(N )
complex floating point operations.
</PRE>
<A name="Comments"><B><FONT SIZE="+1">Further Comments</FONT></B></A>
<PRE>
None
</PRE>
<A name="Example"><B><FONT SIZE="+1">Example</FONT></B></A>
<P>
<B>Program Text</B>
<PRE>
None
</PRE>
<B>Program Data</B>
<PRE>
None
</PRE>
<B>Program Results</B>
<PRE>
None
</PRE>
<HR>
<A HREF=support.html><B>Return to Supporting Routines index</B></A></BODY>
</HTML>
|
