Mo Logo [Home] [Lexikon] [Aufgaben] [Tests] [Kurse] [Begleitmaterial] [Hinweise] [Mitwirkende] [Publikationen]

Mathematik-Online-Lexikon:

Unitäre Diagonalisierbarkeit


A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Übersicht

Sei

$\displaystyle A \;:=\; \left(\begin{array}{rrrr} 1 & 2 & 1 & -2 \\
-2 & 1 & 2 ...
...1 &-2 & 1 & 2 \\
2 & 1 &-2 & 1
\end{array}\right)\in\mathbb{C}^{4\times 4}\;.
$

Überprüfe $ A$ auf unitäre Diagonalisierbarkeit. Bestimme gegebenenfalls eine unitäre Matrix $ U\in\mathbb{C}^{4\times 4}$ und eine Diagonalmatrix $ D\in\mathbb{C}^{4\times 4}$ so, daß $ \bar{U}^\mathrm{t} AU=D$ .

Lösung.

Es wird

$\displaystyle \bar{A}^\mathrm{t} A \;=\; \left(\begin{array}{rrrr} 10 & 0 & -6 ...
... 0 & 10 & 0 \\
0 & -6 & 0 & 10
\end{array}\right) \;=\; A\bar{A}^\mathrm{t}\,
$

d.h. $ A$ ist normal und damit unitär diagonalisierbar.

Das charakteristische Polynom ergibt sich zu

\begin{displaymath}
\begin{array}{rcl}
\chi_A(X)
&=& \det\left(\begin{array}{rr...
...} \\
&=& (X-2)^2 (X-4\mathrm{i})(X+4\mathrm{i})\;.
\end{array}\end{displaymath}

Also sind $ 2$ , $ 4\mathrm{i}$ und $ -4\mathrm{i}$ die Eigenwerte von $ A$ .

Die Eigenräume von $ A$ ergeben sich wie folgt.

Es wird

$\displaystyle \mathrm{E}_A(2) \;=\; \operatorname{Kern}\left(\begin{array}{rrrr...
...1\\ 0\\ 1\\ 0\end{pmatrix},\begin{pmatrix}0\\ 1\\ 0\\ 1\end{pmatrix}\rangle\;.
$

Eine Orthonormalbasis von $ \mathrm{E}_A(2)$ ergibt sich zu

$\displaystyle (\begin{pmatrix}1/\sqrt{2}\\ 0\\ 1/\sqrt{2}\\ 0\end{pmatrix},\begin{pmatrix}0\\ 1/\sqrt{2}\\ 0\\ 1/\sqrt{2}\end{pmatrix})\;.
$

Es wird

$\displaystyle \mathrm{E}_A(4\mathrm{i}) \;=\; \operatorname{Kern}\left(\begin{a...
...(\begin{array}{r}\mathrm{i}\\ -1\\ -\mathrm{i}\\ 1\end{array}\right)\rangle\;.
$

Eine Orthonormalbasis von $ \mathrm{E}_A(4\mathrm{i})$ ergibt sich zu

$\displaystyle (\left(\begin{array}{r}\mathrm{i}/2\\ -1/2\\ -\mathrm{i}/2\\ 1/2\end{array}\right))\;.
$

Es wird

$\displaystyle \mathrm{E}_A(-4\mathrm{i}) \;=\; \operatorname{Kern}\left(\begin{...
...(\begin{array}{r}-\mathrm{i}\\ -1\\ \mathrm{i}\\ 1\end{array}\right)\rangle\;.
$

Eine Orthonormalbasis von $ \mathrm{E}_A(-4\mathrm{i})$ ergibt sich zu

$\displaystyle (\left(\begin{array}{r}-\mathrm{i}/2\\ -1/2\\ \mathrm{i}/2\\ 1/2\end{array}\right))\;.
$

Zur Probe verifizieren wir, daß die Eigenräume bezüglich verschiedener Eigenräume in der Tat zueinander orthogonal sind.

Zusammengesetzt erhalten wir also mit

$\displaystyle U \;=\;
\left(\begin{array}{ccrr}
1/\sqrt{2} & 0 & \mathrm{i}/2 &...
...thrm{i}/2 & \mathrm{i}/2 \\
0 & 1/\sqrt{2} & 1/2 & 1/2 \\
\end{array}\right)
$

die unitäre Diagonalisierung

$\displaystyle \bar{U}^\mathrm{t} A U \;=\; \mathrm{diag}(2,2,4\mathrm{i},-4\mathrm{i})\; .
$

(Autoren: Künzer/Martin/Tentler/Wahrheit)

[Verweise]

  automatisch erstellt am 22.  8. 2006