NE 112 Linear algebra for nanotechnology engineering
11.7 Normal operators
Douglas Wilhelm Harder, LEL, M.Math.
[email protected] [email protected] Normal operators
Introduction
• In this topic, we will
– Define what it is for a linear operator to be normal
– Investigate how such linear operators and their adjoints affect
the length of vectors
– See that the null space and range are identical for both a normal
linear operator and its adjoint
2
Normal operators
Definition
• A linear operator A:U → U is said to be normal if it commutes
with its adjoint
AA = A A
* *
3
Normal operators
Non-examples
• This is not normally true
– For example,
0 1
A=
2 3
0 1 0 2 1 3
AA =
T
=
2 3 1 3 3 13
0 2 0 1 4 6
A A=
T
=
1 3 2 3 6 10
4
Normal operators
Examples of normal operators
Theorem
Every self-adjoint and skew-adjoint linear operator is
normal.
Proof:
If A = A*, then AA* = A2 = A*A.
Thus, a self-adjoint operator is normal.
If A* = –A, then AA* = –A2 = A*A.
Thus, a skew-adjoint operator is normal. ▮
5
Normal operators
Examples of normal operators
• As a consequence:
– Every real symmetric matrix is normal
– Every real skew-symmetric matrix is normal
– Every conjugate symmetric matrix is normal
– Every conjugate skew-symmetric matrix is normal
• There are other matrices that are neither symmetric nor skew-
symmetric that are normal
0 1 0 1 1 0
A = 0 0 1 B = 0 1 1 = A + I3
1 0 0 1 0 1
6
Normal operators
Consequences
Theorem
If A is normal, then ||Au||2 = ||A*u||2 for all u.
Proof: Au
2
= Au, Au
2
= u, A* Au
= u, AA*u
= A*u, A*u
2
= Au*
2
Thus, ||Au||2 = ||A*u||2 . ▮ 7
Normal operators
Consequences
• Note that this is not true in general:
– For example,
0 1
A=
2 3
0 1 1 1 1
= = 26
2 3 1 5 5 2
0 2 1 2 2
= = 20
1 3 1 4 4 2
8
Normal operators
Equality of null space and range
Theorem
If A is a normal linear operator,
then null(A) = null(A*) and range(A) = range(A*).
Proof:
If Au = 0, then Au 2 = 0 .
From our last theorem, it follows that A*u =0
2
Thus, A*u = 0.
Therefore, the null spaces are equal.
9
Normal operators
Equality of null space and range
Theorem
If A is a normal linear operator,
then null(A) = null(A*) and range(A) = range(A*).
Sketch of the proof:
If u ∈ null(A) = null(A*), then Au = A*u = 0
Thus, Au, v = A*u, v = 0 for all v.
**
Thus, u, A v = u, A v = u, Av = 0 for all v.
*
Thus, each vector in the null space is perpendicular to
all vectors each of the ranges.
But dim(null(A)) + dim(range(A)) = n and the same for A*,
thus, these to ranges must be equal. ▮
10
Normal operators
Equality of null space and range
• Let’s look at this normal matrix:
−1 1 0 −1 0 1
A= 0 −1 1 AT = 1 −1 0
1 0 1 −1
0 −1
−1 1 0 0 −1 0 1 0
( A 0 ) ~ 0 −1 1 0
(A
T
0)~ 0 −1 1 0
0 0 0 0 0 0 0 0
−1 1 1 −1 0 1
0 , −1 1
1
, −1
1
1 0 1 0 1 1
11
Normal operators
Equality of null space and range
• This matrix is not normal:
−1 1 0 −1 1 0
A= 1 −1 1 AT = 1 −1 0
0 0 1 −1
0 −1
−1 1 0 0 −1 1 0 0
( A 0 ) ~ 0 0 1 0
(A
T
0)~ 0 1 −1 0
0 0 0 0 0 0 0 0
−1 0 1 −1 1 1
0 , 1 1
1
, −1
1
1 −1 0 0 1 1
12
Normal operators
Summary
• Following this topic, you now
– Understand what it means for a linear operator to be normal:
AA* = A*A
– Understand that all self-adjoint and skew-adjoint linear
operators are normal, but there are others
– Know that for normal matrices, ||Au||2 = ||A*u||2
– Know that if a linear operator is normal,
then the null space and the range of both A and A* are identical
13
Normal operators
References
[1] https://en.wikipedia.org/wiki/Normal_matrix
[2] https://en.wikipedia.org/wiki/Normal_operator
14
Normal operators
Acknowledgments
None so far.
15
Normal operators
Colophon
These slides were prepared using the Cambria typeface. Mathematical equations
use Times New Roman, and source code is presented using Consolas.
Mathematical equations are prepared in MathType by Design Science, Inc.
Examples may be formulated and checked using Maple by Maplesoft, Inc.
The photographs of flowers and a monarch butter appearing on the title slide and
accenting the top of each other slide were taken at the Royal Botanical Gardens in
October of 2017 by Douglas Wilhelm Harder. Please see
https://www.rbg.ca/
for more information.
16
Normal operators
Disclaimer
These slides are provided for the NE 112 Linear algebra for
nanotechnology engineering course taught at the University of
Waterloo. The material in it reflects the authors’ best judgment in
light of the information available to them at the time of preparation.
Any reliance on these course slides by any party for any other
purpose are the responsibility of such parties. The authors accept
no responsibility for damages, if any, suffered by any party as a
result of decisions made or actions based on these course slides for
any other purpose than that for which it was intended.
17
