File: node128.html

package info (click to toggle)
lapack 3.0.20000531a-28
  • links: PTS
  • area: main
  • in suites: sarge
  • size: 61,920 kB
  • ctags: 46,200
  • sloc: fortran: 584,835; perl: 8,226; makefile: 2,331; awk: 71; sh: 45
file content (271 lines) | stat: -rw-r--r-- 8,045 bytes parent folder | download | duplicates (4)
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
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<!--Converted with LaTeX2HTML 98.2 beta6 (August 14th, 1998)
original version by:  Nikos Drakos, CBLU, University of Leeds
* revised and updated by:  Marcus Hennecke, Ross Moore, Herb Swan
* with significant contributions from:
  Jens Lippmann, Marek Rouchal, Martin Wilck and others -->
<HTML>
<HEAD>
<TITLE>Representation of Orthogonal or Unitary Matrices</TITLE>
<META NAME="description" CONTENT="Representation of Orthogonal or Unitary Matrices">
<META NAME="keywords" CONTENT="lug_l2h">
<META NAME="resource-type" CONTENT="document">
<META NAME="distribution" CONTENT="global">
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<LINK REL="STYLESHEET" HREF="lug_l2h.css">
<LINK REL="previous" HREF="node121.html">
<LINK REL="up" HREF="node109.html">
<LINK REL="next" HREF="node129.html">
</HEAD>
<BODY >
<!--Navigation Panel-->
<A NAME="tex2html6013"
 HREF="node129.html">
<IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next"
 SRC="next_motif.png"></A> 
<A NAME="tex2html6007"
 HREF="node109.html">
<IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up"
 SRC="up_motif.png"></A> 
<A NAME="tex2html6003"
 HREF="node127.html">
<IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous"
 SRC="previous_motif.png"></A> 
<A NAME="tex2html6009"
 HREF="node1.html">
<IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents"
 SRC="contents_motif.png"></A> 
<A NAME="tex2html6011"
 HREF="node152.html">
<IMG WIDTH="43" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="index"
 SRC="index_motif.png"></A> 
<BR>
<B> Next:</B> <A NAME="tex2html6014"
 HREF="node129.html">Installing LAPACK Routines</A>
<B> Up:</B> <A NAME="tex2html6008"
 HREF="node109.html">Documentation and Software Conventions</A>
<B> Previous:</B> <A NAME="tex2html6004"
 HREF="node127.html">Real Diagonal Elements of</A>
 &nbsp <B>  <A NAME="tex2html6010"
 HREF="node1.html">Contents</A></B> 
 &nbsp <B>  <A NAME="tex2html6012"
 HREF="node152.html">Index</A></B> 
<BR>
<BR>
<!--End of Navigation Panel-->

<H1><A NAME="SECTION03540000000000000000"></A><A NAME="secorthog"></A>
<BR>
Representation of Orthogonal or Unitary Matrices
</H1>

<P>
A real orthogonal or complex unitary matrix (usually denoted <B><I>Q</I></B>) is often 
represented<A NAME="19987"></A> in
LAPACK as a product of <B>elementary reflectors</B> -- also referred to as
<A NAME="19989"></A>
<A NAME="19990"></A>
<B>elementary Householder matrices</B> (usually denoted <B><I>H</I><SUB><I>i</I></SUB></B>). For example,
<BR><P></P>
<DIV ALIGN="CENTER">

<!-- MATH
 \begin{displaymath}
Q = H_{1} H_{2} \ldots H_{k}.
\end{displaymath}
 -->


<IMG
 WIDTH="133" HEIGHT="30" BORDER="0"
 SRC="img938.png"
 ALT="\begin{displaymath}Q = H_{1} H_{2} \ldots H_{k}. \end{displaymath}">
</DIV>
<BR CLEAR="ALL">
<P></P>
Most users need not be aware
of the details, because LAPACK routines are provided to work with this
representation:

<P>

<UL><LI>routines whose names begin SORG- (real) or CUNG- (complex) can generate
all or part of <B><I>Q</I></B> explicitly;

<P>

<LI>routines whose name begin SORM- (real) or CUNM- (complex) can multiply
a given matrix by <B><I>Q</I></B> or <B><I>Q</I><SUP><I>H</I></SUP></B> without forming <B><I>Q</I></B> explicitly.

<P>

</UL>

<P>
The following further details may occasionally be useful.

<P>
An elementary reflector (or elementary Householder matrix) <B><I>H</I></B> of order
<B><I>n</I></B> is a
unitary matrix<A NAME="19999"></A> of the form
<A NAME="20000"></A>
<A NAME="20001"></A>
<BR>
<DIV ALIGN="RIGHT">


<!-- MATH
 \begin{equation}
H = I - \tau v v^{H}
\end{equation}
 -->

<TABLE WIDTH="100%" ALIGN="CENTER">
<TR VALIGN="MIDDLE"><TD ALIGN="CENTER" NOWRAP><A NAME="Hdef"></A><IMG
 WIDTH="109" HEIGHT="28" BORDER="0"
 SRC="img939.png"
 ALT="\begin{displaymath}
H = I - \tau v v^{H}
\end{displaymath}"></TD>
<TD WIDTH=10 ALIGN="RIGHT">
(5.1)</TD></TR>
</TABLE>
<BR CLEAR="ALL"></DIV><P></P>
where <IMG
 WIDTH="14" HEIGHT="16" ALIGN="BOTTOM" BORDER="0"
 SRC="img220.png"
 ALT="$\tau$">
is a scalar, and <B><I>v</I></B> is an <B><I>n</I></B>-vector, with

<!-- MATH
 $| \tau | ^2   || v || _2 ^2 = 2 \rm {Re}(\tau$
 -->
<IMG
 WIDTH="136" HEIGHT="37" ALIGN="MIDDLE" BORDER="0"
 SRC="img940.png"
 ALT="$\vert \tau \vert ^2 \vert\vert v \vert\vert _2 ^2 = 2 \rm {Re}(\tau$">);  <B><I>v</I></B> is often referred to
as the <B>Householder vector</B><A NAME="20008"></A> .
Often <B><I>v</I></B> has several leading or trailing zero elements, but for the
purpose of this discussion assume that <B><I>H</I></B> has no such special structure.

<P>
There is some redundancy in the representation (<A HREF="node128.html#Hdef">5.1</A>), which can be 
removed in
various ways. The representation used in LAPACK (which differs from
those used in LINPACK or EISPACK) sets <B><I>v</I><SUB>1</SUB> = 1</B>; hence <B><I>v</I><SUB>1</SUB></B> need not
be stored. In real arithmetic, 
<!-- MATH
 $1 \leq \tau \leq 2$
 -->
<IMG
 WIDTH="78" HEIGHT="30" ALIGN="MIDDLE" BORDER="0"
 SRC="img941.png"
 ALT="$1 \leq \tau \leq 2$">,
except that
<IMG
 WIDTH="46" HEIGHT="16" ALIGN="BOTTOM" BORDER="0"
 SRC="img942.png"
 ALT="$\tau = 0$">
implies <B><I>H</I> = <I>I</I></B>.

<P>
In complex arithmetic<A NAME="20010"></A>, <IMG
 WIDTH="14" HEIGHT="16" ALIGN="BOTTOM" BORDER="0"
 SRC="img220.png"
 ALT="$\tau$">
may be 
complex, and satisfies

<!-- MATH
 $1 \leq \rm {Re}(\tau) \leq 2$
 -->
<IMG
 WIDTH="112" HEIGHT="34" ALIGN="MIDDLE" BORDER="0"
 SRC="img943.png"
 ALT="$1 \leq \rm {Re}(\tau) \leq 2$">
and 
<!-- MATH
 $| \tau - 1 | \leq 1$
 -->
<IMG
 WIDTH="86" HEIGHT="34" ALIGN="MIDDLE" BORDER="0"
 SRC="img944.png"
 ALT="$\vert \tau - 1 \vert \leq 1$">.
Thus a complex <B><I>H</I></B> is
not Hermitian (as it is in other representations), but it is unitary,
which is the important property. The advantage of allowing <IMG
 WIDTH="14" HEIGHT="16" ALIGN="BOTTOM" BORDER="0"
 SRC="img220.png"
 ALT="$\tau$">
to be
complex is that, given an arbitrary complex vector <B><I>x</I></B>, <B><I>H</I></B> can be computed
so that <BR><P></P>
<DIV ALIGN="CENTER">

<!-- MATH
 \begin{displaymath}
H^H x = \beta (1, 0, \ldots , 0)^T
\end{displaymath}
 -->


<IMG
 WIDTH="168" HEIGHT="31" BORDER="0"
 SRC="img945.png"
 ALT="\begin{displaymath}H^H x = \beta (1, 0, \ldots , 0)^T \end{displaymath}">
</DIV>
<BR CLEAR="ALL">
<P></P>
with <I>real</I> <IMG
 WIDTH="15" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
 SRC="img50.png"
 ALT="$\beta$">.
This is useful, for example,
when reducing a complex Hermitian matrix to real symmetric tridiagonal form<A NAME="20013"></A>,
or a complex rectangular matrix to real bidiagonal form<A NAME="20014"></A>.

<P>
For further details, see Lehoucq&nbsp;[<A
 HREF="node151.html#lawn72">79</A>].

<P>
<HR>
<!--Navigation Panel-->
<A NAME="tex2html6013"
 HREF="node129.html">
<IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next"
 SRC="next_motif.png"></A> 
<A NAME="tex2html6007"
 HREF="node109.html">
<IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up"
 SRC="up_motif.png"></A> 
<A NAME="tex2html6003"
 HREF="node127.html">
<IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous"
 SRC="previous_motif.png"></A> 
<A NAME="tex2html6009"
 HREF="node1.html">
<IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents"
 SRC="contents_motif.png"></A> 
<A NAME="tex2html6011"
 HREF="node152.html">
<IMG WIDTH="43" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="index"
 SRC="index_motif.png"></A> 
<BR>
<B> Next:</B> <A NAME="tex2html6014"
 HREF="node129.html">Installing LAPACK Routines</A>
<B> Up:</B> <A NAME="tex2html6008"
 HREF="node109.html">Documentation and Software Conventions</A>
<B> Previous:</B> <A NAME="tex2html6004"
 HREF="node127.html">Real Diagonal Elements of</A>
 &nbsp <B>  <A NAME="tex2html6010"
 HREF="node1.html">Contents</A></B> 
 &nbsp <B>  <A NAME="tex2html6012"
 HREF="node152.html">Index</A></B> 
<!--End of Navigation Panel-->
<ADDRESS>
<I>Susan Blackford</I>
<BR><I>1999-10-01</I>
</ADDRESS>
</BODY>
</HTML>