Howell normal form explained
In linear algebra and ring theory, the Howell normal form is a generalization of the row echelon form of a matrix over
, the ring of integers modulo N. The row spans of two matrices agree if, and only if, their Howell normal forms agree. The Howell normal form generalizes the
Hermite normal form, which is defined for matrices over
.
Definition
A matrix
over
is called to be in
row echelon form if it has the following properties:
be the number of non-zero rows of
. Then the topmost
rows of the matrix are non-zero,
, let
be the index of the leftmost non-zero element in the row
. Then
.
With elementary transforms, each matrix in the row echelon form can be reduced in a way that the following properties will hold:
, the leading element
is a divisor of
,
it holds that
.
If
adheres to both above properties, it is said to be in
reduced row echelon form.
If
adheres to the following additional property, it is said to be in
Howell normal form (
denotes the row
span of
):
be an element of the row span of
, such that
for each
. Then
, where
is the matrix obtained of rows from
-th to
-th of the matrix
.
Properties
For every matrix
over
, there is a unique matrix
in the Howell normal form, such that
. The matrix
can be obtained from matrix
via a sequence of elementary transforms.
From this follows that for two matrices
over
, their row spans are equal if and only if their Howell normal forms are equal.
For example, the matrices
A=\begin{bmatrix}
4&1&0\\
0&0&5\\
0&0&0
\end{bmatrix}, B=\begin{bmatrix}
8&5&5\\
0&9&8\\
0&0&10
\end{bmatrix}
have the same Howell normal form over
:
H=\begin{bmatrix}
4&1&0\\
0&3&0\\
0&0&1
\end{bmatrix}.
Note that
and
are two distinct matrices in the row echelon form, which would mean that their span is the same if they're treated as matrices over some field. Moreover, they're in the
Hermite normal form, meaning that their row span is also the same if they're considered over
, the
ring of integers.
However,
is not a field and over general rings it is sometimes possible to nullify a row's pivot by multiplying the row with a scalar without nullifying the whole row. In this particular case,
3 ⋅ \begin{bmatrix}4&1&0\end{bmatrix}\equiv\begin{bmatrix}0&3&0\end{bmatrix}\pmod{12}.
It implies
\begin{bmatrix}0&3&0\end{bmatrix}\inS(A)
, which wouldn't be true over any field or over integers.
References
Bibliography