International

OPEN

Journal

ACCESS

Of Modern Engineering Research (IJMER)

On Normal and Unitary Polynomial Matrices

G. Ramesh1, P. N. Sudha2, R. Gajalakshmi3
1
2, 3

(Department of Mathemtics, Govt Arts College (Auto), Kumbakonam,Tamil Nadu, India)

(Department of Mathemtics, Periyar Maniammai University, Thanjavur, Tamil Nadu, India

ABSTRACT: The properties of polynomial hermitian, polynomial normal and polynomial unitary
matrices are discussed. A characterization for polynomial normal matrix is obtained.
Keywords: Hermitian, normal, unitary matrices, polynomial hermitian, polynomial normal, polynomial
unitary matrices.

I. INTRODUCTION
In matrix theory, we come across some special types of matrices and two among these are normal
matrix and unitary matrix. The normal matrix plays an important role in the spectral theory of rectangular
matrices and in the theory of generalized inverses. In 1918, the concept of normal matrix with entries from the
complex field was introduced by O. Toeplitz who gave a necessary and sufficient condition for a complex
matrix to be normal. Unitary matrices have significant importance in quantum mechanics because they
preserve norms, and thus, probability amplitudes.
In this paper we have introduced polynomial normal matrix and polynomial unitary matrix. Some
properties about these matrices are discussed.
Definition 1.1[1]
A matrix

A M n is said to be normal if A* A A A* , where A* is the complex conjugate transpose of

A.
Theorem 1.2[1]
If A and B are normal and AB = BA then AB is normal.
Definition 1.3[2]
Let A C n x n . The matrix B is said to be unitarily equivalent to A if there exists an unitary matrix U such
that B U AU .
Theorem 1.4 [3]
If A M n is normal if and only if every matrix unitarily equivalent to A is normal.
*

Theorem 1.5 [2]

A is normal if and only if it is unitarily similar to a diagonal matrix.
Definition 1.6[2]
A complex matrix

U M n is unitary if U *U I .

Theorem 1.7 [2]

An n x n complex matrix A is unitary if and only if row (or column) vectors form an orthonormal set
in Cn.
Theorem 1.8[2]

U is unitary if and only if U is unitary.

Theorem 1.9[2]
Determinant value of unitary matrix is unity.
Theorem 1.10[2]
If U M n is unitary then it is diagonalizable.
Definition 1.11[4]

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 5 | Iss.2| Feb. 2015 | 32|

On Normal and Unitary Polynomial Matrices

A Polynomial matrix is a matrix whose elements are polynomials.
Example 1.12

1 x 2
1 x

Let A( x)

x x2
= A2 2 A1 A0 be a polynomial matrix.
2
2 x x
II. Polynomial Normal Matrices

Definition 2.1
A polynomial normal matrix is a polynomial matrix whose coefficient matrices are normal matrices.
Example 2.2

i2 i 2i
A( ) 2
i 2i i
i i 2 i

=
i 2i
2i

i2 2i i

2i2 3i i
2i
2i i

3i
i i

= A2 A1 A0 , where
2

A0 , A1 and A2 are normal matrices.

Some results on Polynomial normal matrices:

Theorem 2.4
If A( ) and B( ) are polynomial normal matrices and A( ) B( ) B( ) A( ) then A( ) B( ) is a
polynomial normal matrix.
Proof
n
n
Let A( ) A0 A1 ............ An and B( ) B0 B1 ............ Bn be polynomial

A0 , A1 , A2 ,....., An and B0 , B1 , B2 ,....., Bn are normal matrices. And also given

A( ) B( ) B( ) A( ) .
A( ) B( ) A0 BO ( A0 B1 A1 B0 ) ......... A0 Bn A1 Bn 1 .......... An B0 n

normal matrices,

B( ) A( ) B0 AO ( B0 A1 B1 A0 ) ......... B0 An B1 An 1 .......... Bn A0 n .

Here each coefficients of

i.e.

and constants terms are equal.

A0 BO B0 AO

A0 B1 A1 B0 = B0 A1 B1 A0 A0 B1 = B0 A1 and A1 B0 = B1 A0 ..
A0 Bn A1Bn 1 .......... An B0 = B0 An B1 An 1 .......... Bn A0
An BO B0 An , A1 Bn 1 B1 An 1 ,., A0 Bn Bn AO .
Now we prove A( ) B( ) is normal.
[ A( ) B( )][ A( ) B( )] A( ) B( )[ A( )][ B( )]
A( ) [ A( )] B( )[ B( )]
[ A( )] A( )[ B( )] B( )
[ A( )][ B( )] A( ) B( )
[ A( ) B( )][ A( ) B( )]
Hence A( ) B( ) is normal.
Example 2.5

i2 2i
2i2 i

Let A( )

| IJMER | ISSN: 22496645 |

0 i
2i2 i i 2i 2 2i 0
=

i
i 2i 2i
0 2i
i 0

www.ijmer.com

| Vol. 5 | Iss.2| Feb. 2015 | 33|

On Normal and Unitary Polynomial Matrices

A2 2 A1 A0 be polynomial normal matrix, where A0 , A1 and A2 are normal

2i i
2i2 i

2i2 i 2i i

0 2i 2 2i 0
i i

=
2i 0
0 2i
i i
B2 2 B1 B0 be polynomial normal matrix, where B0 , B1 and B2 are normal

matrices and let B( )

matrices and

44 23 52 2 1

24 2 4 1
.
A( ) B( ) = B( ) A( ) =
4
2
4
3
2

Now we have to prove A( ) B( ) is normal.

That is, A( ) B( )][ A( ) B( )] = [ A( ) B( )] [ A( ) B( )]

208 167 406 525 304 83

82 4 2

Hence A( ) B( ) is normal.

8
7
6
5
4
3
20 16 40 52 30 8

82 4 2

Theorem 2.6
If A( ) is a polynomial normal matrix if and only if every polynomial matrix unitarily equivalent to A( )
is polynomial normal matrix.
Proof

A( ) is polynomial normal matrix and B( ) U ( )* A( )U ( ), where U ( ) is

polynomial unitary matrix. Now we show that B( ) is polynomial normal matrix.
B( ) B( ) (U ( ) A( )U ( )) (U ( ) A( )U ( ))

= U ( ) A( ) A( )U ( )

= U ( ) A( ) A( ) U ( )
(U ( ) A( )U ( )) (U ( ) A( )U ( ))

= B( ) B( )
Hence B( ) is polynomial normal matrix.
Conversely, assume B( ) is polynomial normal matrix. To prove A( ) is polynomial normal
Suppose

matrix. To prove A

is polynomial

normal matrix

B( ) (U ( ) A( )U ( )) is normal

(U ( ) A( )U ( )) (U ( ) A( )U ( )) (U ( ) A( )U ( )) (U ( ) A( )U ( ))
U ( ) A( )(U ( ) U ( ) ) A( )U ( ) U ( ) A( ) (U ( ) U ( ) ) A( )U ( )
U ( ) A( ) A( ) (U ( ) U A( ) A U ( )

Pre multiply by U ( ) and post multiply by U ( ) , we get

A( ) A( ) A( ) A
Hence A( ) is a polynomial normal matrix.

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 5 | Iss.2| Feb. 2015 | 34|

On Normal and Unitary Polynomial Matrices

Result 2.7
m
m
Let A A0 A1 ..... Am and B B0 B1 ..... Bm be polynomial matrix, where

Ais, Bis Mn(C), i = 0,1,2,..m are normal matrices and A 0B0 =A1B1 =.AmBm= 0. Then
A0*B0 =A1*B1 =.Am* Bm= 0.
Proof
2
m
Let A A0 A1 A2 ..... Am is polynomial normal matrix , where A0, A1....... Am are
normal matrices. And B B0 B1 B2 ..... Bm is a polynomial matrix and also given A0B0
2

=A1B1 =.AmBm= 0. We know that If A is normal and AB =0 then A B = 0.

From this result A0*B0 =A1*B1 =.Am* Bm= 0.

III. Polynomial Unitary Matrices

Definition 3.1
A Polynomial unitary matrix is a Polynomial matrix whose coefficient matrices are unitary matrices.
Definition 3.2
Let A , B M n C where Ai ' s, Bi ' s M n C i= 1, 2, ., m. The polynomial

is said to be unitarily equivalent to A if there exists a polynomial unitary matrix U such

that B = U A U .
matrix B

Theorem 3.3
If U
Proof

is a Polynomial unitary matrix

if and only if U

Let U A0 A1 A2 ..... An
2

is Polynomial unitary matrix.

matrix. Here Ais are unitary matrices.

be polynomial

(ie) A0 A 0 I , A1 A 1 I ,............ An An I .
*

is polynomial unitary matrix

*
*
*
*
n
From (1), U A0 A 1 ..... A n We know that A s are unitary matrices

Hence U is Polynomial unitary matrix . Similarly we can prove the converse.

To prove U

Remark: 3.4
Since all the coefficient matrices of a polynomial unitary matrix is unitary their determinant value is one.
Theorem:3.5
If U A0 A1 A2 ..... An
2

is polynomial unitary matrix, where

Ai M n F and

A0 , A1, A2 ,.... An are unitary matrices then Ais are diagonalizable.

Proof

Given U A0 A1 ..... An is a polynomial unitary matrix, where A0, A1....... An are unitary
n

matrices. By the result 1.17, we have unitary matrices are diagonalizable.

Therefore A0, A1....... An are diagonalizable.
Result: 3.6

If A is a polynomial unitary matrix then the absolute value of all the eigen values of the coefficient
matrices are unity.

IV. Polynomial Normal And Polynomial Unitary Matrices

Theorem 4.1
Polynomial unitary matrices are polynomial normal matrices.
Proof
We know that unitary matrices are normal. In polynomial unitary matrix all the coefficient matrices are
unitary. Hence by the definition 2.1, we get polynomial unitary matrices are polynomial normal matrices.
Definition 4.2
| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 5 | Iss.2| Feb. 2015 | 35|

On Normal and Unitary Polynomial Matrices

A polynomial hermitian matrix is a polynominal matrix whose coefficients are hermitian matrices.
Theorem 4.3

Polynomial hermitian matrix is polynomial normal matrix if the coefficient matrices satisfy AA = A =

A A .
Proof

We know that hermitian matrix with AA = A = A A is normal. In polynomial hermitian matrix

we have all the coefficient matrices are hermitian. By our hypothesis we have coefficient matrices satisfy the

condition AA = A = A A . Now we get all the coefficient matrices are normal.

Hence the theorem.

V. Conclusion
In this paper some of the properties of polynomial hermitian, polynomial normal and polynomial
unitary matrices are derived. Similarly we can extend all the properties of hermitian, normal and unitary
matrices to polynomial hermitian, normal and unitary matrices.

REFERENCES
[1]
[2]
[3]
[4]

A.K.Sharma, Text book of Matrix, Robert Grone, Charles R. Johnson, Eduardo M. Sa, Henry Wolkowicz, Linear
Algebra and its Applications, Volume 87, March 1987, Pages 213225, Normal matrices.
R.A. Horn, C.R. Johnson, Matrix Analysis (Cambridge University Press,Cambridge, 1990).
H. Schwerdtfeger, Introduction to linear algebra and the theory of matrices (P. Noordhoff, Groningen, 1950).
G.Ramesh, P.N.Sudha, On the Determinant of a Product of Two Polynomial Matrices IOSR Journal of
Mathematics(IOSR-JM) Vol.10, Issue 6, Ver.II (Nov-Dec.2014),PP 10-13.

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 5 | Iss.2| Feb. 2015 | 36|