Geometry of Pseudospheres I.

T.H.Marshall

Abstract

The $n$-dimensional pseudospheres are the surfaces in ${bf
R}^{n+1}$ given by the equations ${x_1}^2+{x_2}^2+ ldots
{x_k}^2-{x_{k+1}}^2- cdots - {x_{n+1}}^2=1$ ($1 leq k
leq n+1$).

We consider the pseudospheres as surfaces in $E_{n+1,k}$,
where $E_{m,k}={bf R}^k times {(i{bf R})}^{m-k}$, and
investigate their geometry in terms of the linear algebra of
these spaces. Each of the spaces $E_{m,k}$ has a natural (not
generally positive definite) metric, which is inherited
by the pseudospheres.

We prove that each matrix with columns in
$E_{m,k}$ can be put into a canonical form by
premultiplying by an orthogonal matrix (a
matrix which effects an isometry of $E_{m,k}$).
We term a matrix in this form {em bitriangular}.
This generalizes upper triangular form for real square
matrices.

Keywords

Math Review Classification
Primary 15A21, 51M10

Last Updated
21/3/97

Length
21 pages

Availability
This article is available in: