Algebra 6.178 has four paramaters $x,y,z,t$ taking all integer values,
subject to $A=\left( \begin{array}{cc}
t & x \\
y & z% \end{array}% \right) $ being non-singular modulo $p$. Two such parameter matrices $A$ and
$B$ define isomorphic algebras if and only if% \[
B=\frac{1}{\det P}PAP^{-1}\func{mod}p \]%
for some matrix $P$ of the form% \begin{equation} \left( \begin{array}{ll} \alpha & \beta\\ \omega\beta & \alpha \end{array}% \right) \text{ or }\left( \begin{array}{ll} \alpha & \beta\\
-\omega\beta & -\alpha \end{array}% \right) \end{equation}%
which is non-singular modulo $p$. (Here, as elsewhere, $\omega $ is a
primitive element modulo $p$.) So we need to compute the orbits of GL$(2,p)$
under the action of the subroup of GL$(2,p)$ consisting of matrices of the
form (1). The set of all matrices $P$ of this form is a group $G$ of order $%
2(p^{2}-1)$. The number of orbits is $p^{2}+(p+1)/2-\gcd (p-1,4)/2$.
We show that every orbit contains a matrix $\left( \begin{array}{cc}
0 & x \\
y & z% \end{array}% \right) $ or $\left( \begin{array}{cc}
1 & x \\
y & z% \end{array}% \right) $.
Let $A=\left( \begin{array}{cc}
t & x \\
y & z% \end{array}% \right) $.
If $P=\left( \begin{array}{ll} \alpha & 0 \\
0 & \alpha \end{array}% \right) $ then $\frac{1}{\det P}PAP^{-1}=\allowbreak\left( \begin{array}{cc} \frac{t}{\alpha ^{2}} & \frac{x}{\alpha ^{2}} \\ \frac{y}{\alpha ^{2}} & \frac{z}{\alpha ^{2}}% \end{array}% \right) $. This implies that we can take $t=0$ or $1$ provided $t$ is a
square.
If $P=\left( \begin{array}{ll} \alpha & 0 \\
0 & -\alpha% \end{array}% \right) $ then $\frac{1}{\det P}PAP^{-1}=\allowbreak\left( \begin{array}{cc}
-\frac{t}{\alpha ^{2}} & \frac{x}{\alpha ^{2}} \\ \frac{y}{\alpha ^{2}} & -\frac{z}{\alpha ^{2}}% \end{array}% \right) $, which means that you can take $t=0$ or 1 unless $-1$ is a square,
i.e. unless $p=1\func{mod}4$.
If $P=\left( \begin{array}{ll}
0 & \beta\\ \omega\beta & 0% \end{array}% \right) $ then $\frac{1}{\det P}PAP^{-1}=\allowbreak\left( \begin{array}{cc}
-\frac{z}{\beta ^{2}\omega } & -\frac{y}{\beta ^{2}\omega ^{2}} \\
-\frac{x}{\beta ^{2}} & -\frac{t}{\beta ^{2}\omega }% \end{array}% \right) $, so in the case $p=1\func{mod}4$ you can take $t=0$ or $1$
provided $t$ is a square or $z$ is not a square.
More generally, if $P=\left( \begin{array}{ll} \alpha & \beta\\ \omega\beta & \alpha \end{array}% \right) $ then \begin{eqnarray*}
&&\frac{1}{\det P}PAP^{-1} \\
&=&\frac{1}{\left( \alpha ^{2}-\beta ^{2}\omega\right) ^{2}}\left( \begin{array}{cc}
t\alpha ^{2}+y\alpha\beta -x\alpha\beta\omega -z\beta ^{2}\omega &
x\alpha ^{2}-y\beta ^{2}-t\alpha\beta +z\alpha\beta\\
y\alpha ^{2}-x\beta ^{2}\omega ^{2}+t\alpha\beta\omega -z\alpha\beta \omega & z\alpha ^{2}-y\alpha\beta -t\beta ^{2}\omega +x\alpha\beta \omega \end{array}% \right) . \end{eqnarray*}%
$\allowbreak $So to show that we can take $t=0$ or $1$ even in the case $p=1% \func{mod}4$, we need to show that whatever the values of $t,x,y,z$ we can
always find $\alpha ,\beta $ (not both zero) such that
\[
t\alpha ^{2}+y\alpha\beta -x\alpha\beta\omega -z\beta ^{2}\omega \]%
is a square. Clearly this is possible if $t$ is a square, or if $z$ is not a
square. So let $p=1\func{mod}4$, and assume that $t$ is not a square and
that $z$ is a square. We show that we can always find some value of $\alpha $
for which% \[
t\alpha ^{2}+y\alpha -x\alpha\omega -z\omega \]%
is a square. (Since $z$ is a square, this value of $\alpha $ cannot be
zero.) Completing the square, we have% \[
t\alpha ^{2}+y\alpha -x\alpha\omega -z\omega =t(\alpha +\frac{y-x\omega }{2t%
})^{2}-\frac{(y-x\omega )^{2}}{4t}-z\omega . \]%
Setting $\frac{(y-x\omega )^{2}}{4t}+z\omega $ equal to $\lambda $, we see
that finding $\alpha $ such that $t\alpha ^{2}+y\alpha -x\alpha\omega
-z\omega $ is a square is equivalent to finding $\alpha $ such that \[
t\alpha ^{2}-\lambda \]%
is a square. If $\lambda $ is a square then (since $p=1\func{mod}4$) we see
that $t\alpha ^{2}-\lambda $ is a square when $\alpha =0$. On the other hand
if $\lambda $ is not a square then (since $t$ is not a square) we can find $% \alpha $ such that $t\alpha ^{2}-\lambda =0$.
So we can assume that $t=0$ or 1, This means that we can find
representatives for the $p^{2}+(p+1)/2-\gcd (p-1,4)/2$ orbits in work of
order $p^{5}$. Not brilliant --- it would be nice to do better.
\end{document}
Messung V0.5
¤ Dauer der Verarbeitung: 0.21 Sekunden
(vorverarbeitet)
¤
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.