KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

B.Tech.(IIsem), 2024-25

LINEAR ALGEBRA FOR ENGINEERS (MA103L)

UNIT- 1(MATRIX ALGEBRA)

SYLLABUS:Introduction to Real and Complex Matrices, Elementary Transformation, Rank of

a Matrix by Echelon Form, Solution of System of linear equations by direct method (Gauss
Elimination Method), Solution of System of linear equations by Iterative method (Gauss Seidal
Method), Linear dependence and independence of vectors.

CONTENTS

S.No MATRICES PAGE

NO.
1.1 Introduction 2
1.2 Types of Matrices 2
1.3 Basic Operations on Matrices 7
1.4 Transpose of a Matrix 9
1.5 Symmetric Matrix 10
1.6 Skew-Symmetric Matrix 10
1.7 Complex Matrices 11
1.8 Hermitian and Skew-Hermitian matrix 13
1.9 Unitary Matrix 13
1.10 Elementary Transformations (or Operations) 19
1.11 Rank of a Matrix by Echelon Form 20
1.12 Solution of System of Linear Equations(Gauss Elimination) 24
1.13 Solution of System of Linear Equations(Gauss Seidal) 38
1.14 Linear Dependence and Independence of Vectors 44

1
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

MATRICES

1.1 INTRODUCTION
The term ‘Matrix’ was given by J.J.Sylvesterabout1850,but was introduced first by Cayley in 1860. By a
‘matrix’ we mean “rectangular array” of numbers. Matrices (plural of matrix) find applications in solution
of system of linear equations, probability, mathematical economics, quantum mechanics, electrical
networks, curve fitting, transportation problems etc. Matrices are easily agreeable for computers.

Matrix inverse can be used in Cryptography. It can provide a simple and effective procedure for encoding
and decoding messages.

Matrix

A rectangular array of m. n numbers (real or complex) arranged in m rows (horizontal lines) and n columns
(vertical lines) and enclosed in brackets [ ] is called a matrix of order m n .It is also called m n
matrix.

The numbers in the matrix are called entries or elements of the matrix.

Elements of a matrix are located by the double subscript ij where i denotes the row and j the column.
 a11 a12 ... a1n 
a a22 ... a2 n 
 21
The matrix A is written as A =  ... ... ... ... 
 
am1 am 2 ... amn  mn

If all these entries are real, then the matrix A is called a real matrix.

Note: Matrix has no numerical value.

1.2 TYPES OF MATRICES

(a) Square Matrix

In a matrix, if the number of rows = number of columns = n, then it is called a square matrix of order n.

3 4 5 
 
Example : A = 5 6 7  is a square matrix since its order is 3  3 .
9 3 2 

(b) Row Matrix

A matrix with only one row is called a row matrix.

Examples:

2
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Let A = a11a12 a13a14 .............a1n  . It is a row matrix with n columns. So, it is a row matrix of
type 1  n .

Also, A = 4 5 6 7 . It is a row matrix with 4 columns. So, it is a row matrix of type 1 4 .

(c) Column Matrix

A matrix with only one column is called a column matrix.

Examples:

 a11 
a 
 21 
A =  a31 
Let
 
 . 
 an1 

It is a column matrix with n rows. So, it is a column matrix of type n  1 .

1 
 2
 
Also, Let
A =  3
 
 4
5

It is a column matrix with 5 rows. So, it is a column matrix of type 5  1 .

(d) Diagonal Matrix

A square matrix in which all the non- diagonal elements are zero is called a diagonal matrix

Example:

2 0 0 
 0  is a diagonal matrix of order 3.
Let A = 0 3
0 0 − 4

(e) Scalar Matrix

In a diagonal matrix if all the diagonal elements are equal to a non-zero scalar  , then it is called a
scalar matrix.

Example:

3
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 0 0
  0  is a scalar matrix of order 3.
Let A =  0
 0 0  

(f) Unit Matrix or Identity matrix

In a diagonal matrix if all the diagonal elements are equal to 1, then it is called a unit matrix or
identity matrix.

1 0 0
1 0 
Examples: 1,  0
1 
, 0 1
0 0 0 1

are identity matrices of orders 1,2,3 respectively.They are denoted byI1,I2,I3.

In general, In is the identity matrix of order n .

(g) Zero Matrix or Null matrix

In a matrix(rectangular or square), if all the entries are equal to zero, then it is called a zero matrix or
null matrix.

Example:

0 0 0
0 0
A = 0 0, B = 
0 0
0
0
0 0 0 0 0 0

are zero matrices of type 3 3 , 2 4 .

(h) Triangular Matrix

A square matrix A= [aij] is said to be upper triangular matrix if all the entries below the main diagonal
are zero. That is aij = 0 if i  j.

A square matrix A= [aij]is said to be lower triangular matrix if all the entries above the main diagonal
are zero. That is aij = 0 if i  j.

Examples:

4
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

4 1 0 2 
− 1 2 3 
 1 
4, 
0 2 3
The matrices A =  0 1
0 0 0 − 2 are upper triangular matrices.
 0 0 5  
0 0 0 5 

and

1 0 0
 2 0 
The matrices A =  , 2 − 1 0 are lower triangular matrices.
 − 1 0  0 2 1


(i) Singular and Non-singular Matrices

A square matrix A is said to be singular if A = 0 and non-singular if A  0.

Example:

2 − 3 4
A = 2 − 3 4 is a singular matrix since A = 0 .
0 − 1 1 

(j) Orthogonal Matrix

A square matrix A is called an orthogonal matrix if AA = AA = I .

Note: If A and B are any two orthogonal matrices, AB will also be an orthogonal matrix.

 cos sin  
Example: A = 
− sin  cos 

(k) Elementary matrix

The matrix obtained from a unit matrix I by subjecting it to one of the E-operations is called an
elementary matrix.
1 0 0
Example: Let I = 0 1 0
 
0 0 1
1 0 0
Operating R23 or C23 on I, we get the elementary matrix 0 0 1 .It is denoted by E23.
 
0 1 0

1.3 Basic Operations on Matrices

5
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

(a) Addition and subtraction of Matrices

Two matrices are said to be conformable for addition and subtraction if they have the same order.If A
and B are two matrices of the same order , then their sum A+B is a matrix each element of which is
obtained by adding the corresponding elements of A and B.

Let A= [aij] and B=[bij] be two matrices of the same type m n .Then A+B=[cij] where cij=aij+bij for all
i, j and A + B is of type m n .

Similarly, if A and B are two matrices of the same order , then their difference A-B is a matrix whose
element are obtained by subtracting the elements of B from the corresponding elements of A.

Let A= [aij] and B = [bij] be two matrices of the same type m n .Then A-B = [cij] where

cij=aij-bij for all i, j .

− 1 2 3 1 2 3 
A=
Example :If ,   then
 0 1 5 1 0 − 2
 − 1 + 1 2 + 2 3 + 3  0 4 6 
A+ B =   = 1 1 3
 0 + 1 1 + 0 5 − 2   

We see that A and B are of type 2  3 and A + B is also of the type 2  3

− 1 − 1 2 − 2 3 − 3  − 2 0 0
A− B =  =
 0 −1 1− 0 5 + 2  − 1 1 7

(b) Scalar Multiplication of Matrices

LetA= [aij] be an m n matrix and k be a scalar. Then kA=[kaij].

Example:

a a12 a13   ka ka12 ka13 

A =  11  then kA =  11
ka23 
If .
a21 a22 a23  ka21 ka22

− a − a12 − a13 
− A =  11
− a23 
Henceifk = -1,then .
− a21 − a22

( c) Multiplication of Matrices

If A and B are two matrices such that the number of columns of A is equal to the number of rows of B,
then the product AB is defined. Two such matrices are said to be conformable for multiplication. In the
product AB,A is known as pre-factor and B is known as post-factor.

Let Let A= [aij] be an m  p matrix and B= [bij] be an p  n matrix ,then AB is defined and AB=[cij]is
p
an m n matrix, where cij =  aik bkj .
k =1

6
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

th
That is cij is the sum of the products of the corresponding elements of the i th row of A and the j
column of B.

Example:

1 1 2 1 2
 3 and B = 3 1
Let A = 0 1
2 2 1 2 1

Since A is of type 3  3 and B is of the type 3 2 , AB is defined and AB is of type 3 2 .

1 1 2 1 2 1.1 + 1.3 + 2.2 1.2 + 1.1 + 2.1  8 5

AB = 0 1 3 3 1 = 0.1 + 1.3 + 3.2 0.2 + 1.1 + 3.1 =  9 4
2 2 1 2 1 2.1 + 2.3 + 1.2 2.2 + 2.1 + 1.1 10 7

Note:If A and B are square matrices of order n, then both AB and BA are defined,but not necessarily

equal. That is ,AB ≠BA, in general.So, matrix multiplication is not commutative.

Properties of Addition, Scalar Multiplication and Multiplication

1. If A, B, C are matrices of the same type, then

i. ) A + B = B + A

ii ) A + (B + C ) = ( A + B ) + C

iii) A + 0 = A

iv) A + (− A) = 0

v)  ( A + B ) = A + B

vi) ( +  )A = A + A

vii)  (A) = ( )A for any scalars  ,  .

2. If A, B, C are conformable for multiplication, then

(i)  ( AB) = (A)B + A(B )

(ii) A(BC ) = ( AB)C

(iii) ( A + B )C = AC + BC, where A and B are of type m  p and C is of the type p  n

7
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

n −1
(iv)If A is a square matrix, then A = A  A, A = A  A,.....A = A A
2 3 2 n

1.4 Transpose of a Matrix

Given a matrix A, then the matrix obtained from A by changing its rows into columns and columns into
rows is called the transpose of A and is denoted by AorAt .

Clearly, (i) If the order of A is m n then order of A is n  m.

(ii ) (i, j ) element of A = ( j , i ) element of A.

th
 th

1 2
0 − 1
1 0 2 5
A=  , then A = 
Example: If
2 −1 3 7 2 3
 
5 7

Properties of Transpose of a Matrix

If Aand B denote the transposes of A and B respectively, then

(a) ( A) = A i.e., the transpose of the transpose of a matrix is the matrix itself.
(b) ( A + B ) = A + B i.e., the transpose of the sum of two matrices is equal to the
sum of their transposes.

(c) ( AB) = BA i.e., the transpose of the product of two matrices is equal to the
product of their transposes taken in the reverse order.

1.5 Symmetric Matrix

 
A square matrix A = a ij is said to be symmetric if A = A i.e., if the transpose of the matrix is equal
to the matrix itself.
 
Thus, for a symmetric matrix A = a ij , aij = a ji .

 − 1 2 3  a h g
Example:  2 5 6,  h b f  are symmetric matrices.
  
 3 6 4  g f c 

1.6 Skew-Symmetric Matrix

 
A square matrix A = a ij is said to be symmetric if A = −A i.e., if the transpose of the matrix is equal
to the negative of the matrix.

8
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 
Thus, for a symmetric matrix A = a ij , aij = −a ji .
Putting j=i, aii = −aii  2aii = 0 or aii = 0 for all i. Thus, all diagonal elements of a skew-symmetric
matrix are zero.
0 2 − 3  0 h − g

Example: − 2 0 1 , − h 0 f  are skew-symmetric matrices.

 3 − 1 0   g −f 0 

Theorem: Every square matrix can uniquely be expressed as the sum of a Symmetric

matrix and a Skew-symmetric matrix.

Proof :Let A be any square matrix.

Consider B =
1
( A + A) and C = 1 ( A − A).
2 2

B+C =
1
( A + A) + 1 ( A − A) = A
2 2

B =
1
( A + A) = 1  A + ( A)  = 1 ( A + A) = 1 ( A + A) = B
2 2  2 2

 B is a symmetric matrix.

C =
1
( A − A) = 1  A − ( A)  = 1 ( A − A) = − 1 ( A − A) = −C
2 2  2 2

 C is a skew - symmetric matrix.

Hence every square matrix A can be expressed as A= B+C, B =

1
( A + A) is a symmetric matrix and
2
C=
1
( A − A) is a skew-symmetric matrix.
2

To prove that there is only one way in which A can be expressed as the sum of a symmetric matrix
and a skew-symmetric matrix, suppose

A=P+Q ……….. (1)

is another such representation in which P is a symmetric matrix, and Q is a skew-symmetric matrix.

i.e. P = P and Q = -Q


Then A = (P + Q ) = P + Q = P − Q...............(2)

From (1) and (2) , A + A = 2P and A − A = 2Q.

9
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

⸫ P=
1
( A + A) = B and Q = 1 ( A − A) = C. so that P+Q =B+C
2 2

Hence the result.

1.7 Complex Matrices

A matrix having complex numbers as its elements is called a complex matrix.

 2i −i 2
A=
3
Example:
3 − 2i 0

Complex Conjugate of a Matrix

If A is a matrix having complex numbers as its elements, then the matrix obtained from A by replacing
each element of A by its corresponding complex conjugate is called the conjugate matrix of A and is
denoted by A .
2 + 3i 4 − 6i 3  2 − 3i 4 + 6i 3 
Thus if A =  .then A =  
 1 − i 5 + 2i 5i   1 + i 5 − 2i − 5i 

Some Properties of Complex Matrices

If A and B are two complex matrices, then

(i) (A ) = A
(ii) (A + B) = A + B
(iii) (iii) (kA) = k A where k is a complex number
(iv) (iv) (AB ) = A B

Transposed Conjugate Matrix

The transpose of the conjugate of a matrix A is called transpose conjugate of matrix A is denoted by

A = A . ( )

( )
Clearly the transpose of the conjugate and conjugate of the transpose of matrix are equal i.e., A = ( A).
2 + 3i 5 − 6i   2 − 3i 4 − i 
then A = (A ) = ( A) = 

Thus if A =   3 
.
 4+i 3  5 + 6i

Some Properties of Transposed Conjugate Matrix

(i) A( )
= A.
(ii) If A and B are two matrices conformable for product AB, then

10
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

( AB) = B A .
(iii) ( A + B ) = A + B .
  

(iv) (kA) = kA where k is a scalar quantity.

 

1.8 Hermitian and Skew Hermitian Matrix

A square matrix A is called Hermitian if A = A.

In a Hermitian matrix, the diagonal elements are all real, while every other element is the conjugate
complex of the element in the transposed position.
 5 2 + i − 3i 
 4 3 + 2i 

Examples: A = 2 − i − 3 1 − i  , A = 
  3 − 2i 5  are Hermitian matrix.
 3i 1 + i 0  

A square matrix A is called Skew- Hermitian if A = − A.

In a skew-Hermitian matrix, the diagonal elements are zero or purely imaginary numbers of the form
iβ where β is real. Every other element is the negative of the conjugate complex of the element in the
transposed position.
 3i 1+ i 7 
  , A =  2i 3 + 4i 
Examples : A = − 1 + i − −  5i  are Skew-Hermitian matrix.
 0 2 i 
 − 7 2 − i  − 3 + 4i
−i 

1.9 Unitary Matrix

A square matrix A is said to be unitary if AA = I = A A.

The determinant of a unitary matrix is of unit modulus. For a matrix to be unitary, it must be non-singular.
1  1 1 + i
Example : A =  
3 1 − i − 1  is a Unitary matrix.

Question based on Hermitian, Skew Hermitian Matrix, Unitary matrix

2 + i 3 − 1 + 3i 
Example 1: If A =   , verify that A A is a Hermitian matrix where A
 − 5 i 4 − 2i 
is the conjugate transpose of A.

 2+i −5 
Solution: A =
 i 
 3
− 1 + 3i 4 − 2i 

11
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 2−i −5 

A = ( A) =  3

− i 
− 1 − 3i 4 + 2i 

 2−i −5 
 2 + i 3 − 1 + 3i 

A A= 3 − i  
− 5 i 4 − 2i 
− 1 − 3i 4 + 2i  
 30 6 − 8i − 19 + 17i   30 6 + 8i − 19 + 17i 

=  6 + 8i 10  
− 5 + 5i  = B(say) Now B =  6 − 8i 10 − 5 − 5i 
− 19 − 17i − 5 − 5i 30  − 19 − 17i − 5 − 5i 30 
 30 6 − 8i − 19 + 17i 
 
B = (B) =  6 + 8i

10 − 5 + 5i  = B
− 19 − 17i − 5 − 5i 30 


Hence B = A A isa Hermitian matrix.

Example 2: If A and B are Hermitian, show that AB-BA is skew-Hermitian.

Solution: A and B are Hermitian  A = A and B = B

Now ( AB − BA) = ( AB ) − (BA)
  

B A − A B = BA − AB = −( AB − BA)
 AB − BA is skew-Hermitian.

Example 3: If A is a skew-Hermitianmatrix, then show that iA is Hermitian.

Solution: A is a skew- Hermitian matrix  A = − A

Now (iA) = i A = (− i )(− A) = iA
 

 iA is a Hermitian matrix.

Example 4: Show that every square matrix is expressible as the sum of a Hermitian matrix and Skew-
Hermitian matrix.

Solution: A be any square matrix. Since

(A + A ) = A + (A ) = A + A = A + A
      

and (A − A ) = A − (A ) = A − A = −(A − A )
      

⸫ A + A is Hermitian and A − A is skew-Hermitian.

12
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Now, A =
1
2
( 1
) ( )
A + A + A − A = P + Q (say)
2
where P is Hermitian and Q is skew-Hermitian .Thus,every square matrix can be expressed as the sum
of a Hermitian matrix and a skew-Hermitian matrix.

 2 + 2i 4 5 − 5i 
Example 5 : Express the matrix A =
 3 + i 4 + 2i  as the sum of a Hermitian matrix and
 4i
− 3 + 3i − 4 9 
Skew-Hermitian matrix.
2 − 2i − 4i − 3 − 3i 
Solution: A = ( A) =  4

3−i − 4 

5 + 5i 4 − 2i 9 
Now,

 4 4 − 4i 2 − 8i   2 2 − 2i 1 − 4i 
1
2
(  1
)
A + A = 4 + 4i
2
6  
2i  = 2 + 2i 3 i 
(1)
 2 + 8i − 2i 18  1 + 4i −i 9 
 4i 4 + 4i 8 − 2i   2i 2 + 2i 4 − i 
1
2
(  1
)
A − A = − 4 + 4i
2
2i  
8 + 2i  = − 2 + 2i i 4 + i 
(2)
 − 8 − 2i − 8 + 2i 0   − 4 − i − 4 + i 0 

1
Clearly
2
( )
A + A Hermitian and
1
2
( )
A − A is skew-Hermition.

From (1) and (2), we have

1
2
( 1
A = A + A + A − A
2
) ( )
 2 2 − 2i 1 − 4i   2i 2 + 2i 4 − i 

A = 2 + 2i 3  
i  + − 2 + 2i i 4 + i 
1 + 4i −i 9   − 4 − i − 4 + i 0 

1  1 1 + i
Example 6: Show that the matrix A =   is unitary.
3 1 − i − 1 
1  1 1 + i
Solution: Given A =  
3 1 − i − 1 
1  1 1 + i
 A = ( A) =  
3 1 − i − 1 
 1  1 1 + i   1 1 + i  1 3 0 1 0
Now AA = = = = I.
3 1 − i − 1  1 − i − 1  3 0 3 0 1
Similarly A A = I .

13
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 AA  = I = A A.

Hence A is unitary.
 0 1 + 2i 
is a matrix, then find ( I-N )(I + N ) and show that it is a unitary
−1
Example 7: If N =  
− 1 + 2i 0 
matrix.

Solution: Given
 0 1 + 2i 
N=
− 1 + 2i 0 

Also
1 0 
I= .
0 1 

 1 − 1 − 2i 
I-N = 
1 
Now
1 − 2i
 1 1 + 2i 
I+N =
− 1 + 2i 1 
1 + 2i
I+N =
1
− 1 + 2i 1
( )
= 1 − − 1 + 4i 2 = 6

 1 − 1 − 2i 
Adj (I + N ) = 
1 − 2i 1 
1 1 − 1 − 2i 
Thus (I + N) = Adj(I + N ) = 
−1 1
1 
.
I+N 6 1 − 2i
1 1 − 1 − 2i   1 − 1 − 2i 
Thus (I - N )(I + N ) =
−1

6 1 − 2i  
1  1 − 2i 1 
.

1  − 4 − 2 − 4i 
= = P, (say)
6 2 − 4i − 4 

To prove P is unitary,we shall show Pθ P = I.

1  −4 2 + 4i 
We have P = (P) =
θ

6 − 2 + 4i − 4 
1  −4 2 + 4i   − 4 − 2 − 4i 
Pθ P =
36 − 2 + 4i − 4  2 − 4i
 − 4 
1 36 0  1 0
 Pθ P = = = I.
36  0 36 0 1

14
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Hence P i.e.,(I-N)(I+N)-1 is unitary.

Practice Question:
 + i −  + i 
is unitary if  +  +  +  = 1.
2 2 2 2
1) Show that the matrix  
  + i  − i 
2) If A is any square matrix, prove that A + A , AA , A A are all Hermitian and A− A is skew-
  

Hermitian.
 2 7 - 4i - 2 + 5i 

3) Show that A = 7 + 4i -2 3 + i  is a Hermitian matrix.

- 2 - 5i 3 - i 3 
 i 3 + 2i - 2 − i 

4) Show that A = − 3 + 2i 0 3 − 4i  is a skew-Hermitian matrix.

 2 - i - 3 - 4i - 2i 
 -1 2 + i 5 - 3i 

5) If A = 2 - i 7 5i , Show that Ais a Hermitian matrix and iA is a skew-Hermitian

5 + 3i - 5i 2 
matrix.
1 + i − 1 + i 
6) Prove that A =   is a Unitary matrix.
1 + i 1 − i 
i 0 0
7) Show that A = 0 0 i  is skew-Hemitian matrix and also unitary.

0 i 0
1 + i −1 + i
Prove that A = 
1 − i 
8) is a Unitary matrix.
1 + i
9) If A and B are two unitary matrices,show that AB is a unitary matrix.
10) If A is a unitary matrix, show that AT and A-1 is unitary.

15
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1.10 Elementary Transformations (or Operations)

Any one of the following operations on a matrix is called an elementary transformation (or E-
operation).
(i) Interchange of two rows or two columns.

The interchange of ith and jth columns is denoted by Rij.

The interchange of ith and jth columns is denoted by Cij.

(ii) Multiplication of (each element of) a row or column by a non-zero number k.

The multiplication of ith row by k is denoted by Ri(k).

The multiplication of ith column by k is denoted by Ci(k).

(iii) Addition of k times the elements of a row (or column) to the corresponding elements of another
row(or column),k≠0.

The addition of k times the jth row to the ith row is denoted by Rij(k).

The addition of k times the jth column to the ith column is denoted by Cij(k).

If a matrix B is obtained from a matrix A by one or more E-operations, then B is said to be equivalent
to A. Two equivalent matrices A and B are written as A~B.

1.11 Rank of a Matrix

Let A be any m n matrix. It has square sub-matrices of different orders. The determinants of these
square sub-matrices are called minors of A.

A matrix A is said to be of rank r if

(i) It has at least one non-zero minor of order r.

(ii) All the minors of order (r+1) or higher than r are zero.

Symbolically, rank of A= r is written as  ( A) = r.

Note: (1) If A is a null matrix, then rank of A= 0.

(2)If A is not a null matrix, then rank of A  1 .

(3)If A is a non-singular n  n matrix , then rank of A = n.

(4)Let A be any matrix (square or rectangular).From this matrix A, delete all columns and rows leaving
a certain p columns and p rows. Now if p > 1, then the elements which have been left, constitute a square
matrix of order p. The determinant of this square matrix is called a minor of A of order p.

16
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1.11 Methods of finding Rank of Matrix

(a) Echelon Form Method (Row Reduced Echelon form)

➢ Apply only elementary row operations on A. Then the number of non-zero rows is the
rank of A.
➢ A matrix is reduced in echelon form as
• The first non-zero element in a row should be unity (if possible).
• All the non-zero rows, if any, precede the zero rows.
➢ The rank of the matrix is equal to the number of non-zero diagonal elements when it has
been reduced to echelon form.ie number of non-zero rows is equal to the rank of the matrix.

Determine the rank of the following matrices by reducing to Echelon form:

Example 1 : Determine the rank of the following matrices by reducing to Echelon form:

4 2 3 

A= 8 4 6 
− 2 − 1 − 1.5

Solution: Apply elementary row operations on A

4 2 3
By R21(-2), R31(1/2) ~ 0 0 0
 
0 0 0

The number of non-zero row is one. So the rank of A is one.

1 5 4  1 1 1
Example 2: A = 0 3 2 , B =  2 2 2
   
2 13 10 3 3 3

Find rank of A, rank of B, rank of A+B, rank of AB and rank of BA.

1 5 4 
Solution: A = 0 3 2 
 
2 13 10

1 5 4
By R31(-2) ~ 0 3 2
 
0 3 2

17
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1 5 4
By R32(-1) ~ 0 3 2
 
0 0 0

Rank of A is 2 since the number of non-zero rows is 2.

1 1 1 
B = 2 2 2
3 3 3

1 1 1
By R21(-1), R31(-3), ~ 0 0 0
 
0 0 0

Rank of B is 1 since the number of non-zero rows is 1.

2 6 5 
A + B = 2 5 4 
5 16 13

2 6 5 

By R21(-2), R31(-5/2), ~ 0 − 7 − 6 
 
0 − 1 1 / 2

Rank of A+B is 3 since the number of non-zero rows is 3.

23 23 23
AB = 12 12 12 
58 58 58

1 1 1
By R1(1/23), R21(-12), R31(-58), ~ 0 0 0,
 
0 0 0

Rank of AB is 1 since the number of non-zero rows is 1.

3 21 16 
BA = 6 42 32
9 63 48

1 1 1
By R21(-2), R31(-3), ~ 0 0 0,
 
0 0 0

18
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Rank of BA is 1 since the number of non-zero rows is 1.

Example 3:Determine the rank of the following matrices by reducing to Echelon form:

1 2 −2 3 
2 5 − 4 6 
A= 
 − 1 − 3 2 − 2
 
2 4 −1 6 

1 2 −2 3 
2 5 − 4 6 
Solution: A = 
 − 1 − 3 2 − 2
 
2 4 −1 6 

1 2 − 2 3
0 1 0 0
By R21(-2), R31(1), R41(-2), ~ 
0 − 1 0 1
 
0 0 3 0

1 2 −2 3
0 1 0 0
By R31(1), R34,R3(1/3), ~ 
0 0 1 0
 
0 0 0 1

Rank of A is 4 since the number of non-zero rows is 4.

Example 4: Determine the values of b such that the rank of A is 3.

1 1 − 1 0
4 4 − 3 1
A=
b 2 2 2
 
9 9 b 3

1 1 − 1 0
4 4 − 3 1
Solution: A = 
b 2 2 2
 
9 9 b 3

 1 1 −1 0
 0 0 1 1
By R21(-4), R31(-2), R41(-9), ~ 
b − 2 0 4 2
 
 0 0 b+9 3

19
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 1 1 −1 0
 0 0 1 1 
By R32 (-4), R42(-3) ~ 
b − 2 0 0 − 2
 
 0 0 b+6 0 

 1 1 −1 0
 0 0 1 1 
By R43 ~ 
 0 0 b+6 0 
 
b − 2 0 0 − 2

Cases:(i) If b=2,determinant of A= 0 ,rank of A=3.

(ii)If b = -6, number of non-zero rows is 3,rank of A=3.

Example 5: Reduce A to Echelon form and find rank of A.

1 3 − 1 2
0 11 − 5 3
A= 
2 − 5 3 1 
 
4 1 1 5

Solution: Applying elementary row operations on A

1 3 −1 2 
0 11 − 5 3 
R31 (− 2), R 41 (− 4) ~  
0 − 11 5 − 3
 
0 − 11 5 − 3

1 3 − 1 2
0 11 − 5 3
R32 (1), R 42 (1) ~  
0 0 0 0
 
0 0 0 0

1 3 −1 2 
 −5 3 
 1  0 1
R2   ~ 11 11
 11  0 0 0 0
 
0 0 0 0 

20
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

 4 13 
1 0
11 11 
 −5 3 
R12 (− 3) ~ 0 1 
 11 11 
0 0 0 0
0 0 
 0 0

This is the Echelon form. Rank of A is 2 since there are two non-zero rows.

1.12 Solution of System of Linear Equations (Gauss Elimination Method)

Consider the system of linear equations in three unknown

a1 x + b1 y + c1 z = d1 

a2 x + b2 y + c2 z = d 2 (3 equations in 3 unknowns)
a3 x + b3 y + c3 z = d3 

In matrix notation, these equations can be written as

 a1 x + b1 y + c1 z1   d1 
a x + b y + c z  = d 
 2 2 2   2
 a3 x + b3 y + c3 z   d 3 

 a1 b1 c1   x   d1 
or a2 b2 c2   y  = d 2 
 a3 b3 c3   z   d3 

or AX = B

 a1 b1 c1   x

where A = a2 b2 
c2  is called the co-efficient matrix, X =  y  is the column matrix of

 a3 b3 c3   z 
 d1 
 
unknowns, B = d 2 is the column matrix of constants.
 
 d 3 

HOMOGENEOUS LINEAR EQUATIONS

If all constants d1 = d 2 = d3 = 0 ,then B = 0 and the matrix equation AX = B reduces to

AX = 0 .

Such a system of equations is called a system of homogeneous linear equations.

21
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

NON-HOMOGENEOUS LINEAR EQUATIONS

If atleast one of constants d1 , d 2 , d 3 is non-zero,then B  0. Such a system of equations is called a

system of non-homogeneous linear equations.

Augmented matrix

The matrix  A : B  in which the elements of A (coefficient matrix) and B (Column matrix of
constant) are written side by side is called the augmented matrix.

Example: Consider the system of linearequation

a1 x + b1 y + c1 z = d1 

a2 x + b2 y + c2 z = d 2 (3 equations in 3 unknowns)
a3 x + b3 y + c3 z = d3 

 a1 b1 c1 : d1 

Augmented matrix A : B  = a2 b2 c2 : d 2 

 a3 b3 c3 : d3 

FOR A SYSTEM OF NON-HOMOGENEOUS LINEAR EQUATION AX = B

 
 
(1) Solving the matrix equation AX = B means finding X ,i.e., finding a column matrix  such
 
  
 x   
   
that X = y =  . Then x =  , y =  , z =  .
   
 z    

(2)The matrix equation AX = B need not always have a solution .It may have no solution or a
unique solution or an infinite number of solutions.

(3) A system of equations having no solution is called an inconsistent system of equations.

(4) A system of equations having one or more solution is called a consistent system of equations.

FOR A SYSTEM OF HOMOGENEOUS LINEAR EQUATION AX = 0.

(1) X=0 is always a solution .This solution in which each unknown has the value zero is called the Null

Solution or the Trivial Solution. Thus a homogeneous system is always consistent.A system of

homogeneous linear equation has either the trivial solution or an infinite number of solutions.

(2) If rank (A) = number of unknowns,the system has only the trivial solution.

22
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

(3) If rank(A) <number ofunknowns,the system has an infinite number of non-trivial solutions.

Example 1: Examine the following equations for consistency and if consistent, find the complete solution.

x + y + z + 3 = 0,3x + y − 2 z + 2 = 0,2 x + 4 y + 7 z − 7 = 0.

Solution: Here

x + y + z = −3
3x + y − 2 z = −2
2 x + 4 y + 7 z = 7.

The matrix equation AX=B is given by

1 1 1   x   − 3
3 1 − 2  y  = − 2

2 4 7   z   7 

1 1 1 : −3
A : B = 3 1 − 2 : −2
2 4 7 : 7 
1 1 1 : −3
R21 (− 3), R31 (− 2) ~ 0 −2 − 5 : 7 
0 2 5 : 13 
1 1 1 : −3
R32 (1) ~ 0 −2 − 5 : 7 
0 0 0 : 20
1 1 1 : −3 
 1  1   5 :−7 
R2  − , R3  ~ 0 1
 2  20   2 2
0 0 0 : 1 
Which is in Echelon form.⸫ Rank of [A:B] =3.

Now deleting the last column from the Echelon form of [A:B],we have
1 1 1 

Echelon form of A= 0 1 5 
 2
0 0 0 

Rank of A= 2= Number of non zero rows

⸫Rank of [A:B] =3≠Rank of A=2.

23
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Hence the given systems are inconsistent.

Example 2: Test the consistency and hence solve the following system of equations:

x + 2 y + z = 2,3x + y − 2z = 1,4x − 3 y − z = 3,2x + 4 y + 2z = 4.

Solution: The matrix equation AX=B is given by

 1 −3 2 
A = − 3 3 − 1
−1

 2 − 1 0  (1)

1 2 1 :2
3 1 − 2 :1 
A : B = 
4 − 3 − 1 :3
 
2 4 2 :4
1 2 1 : 2
0 − 5 − 5 : − 5
R21 (− 3), R31 (− 4), R41 (− 2) ~ 
0 − 11 − 5 : − 5
 
0 0 0 : 0
1 2 1 : 2
0 −5 −5 : − 5
R32 (− 1) ~ 
0 −6 0 : 0
 
0 0 0 : 0
1 2 1 : 2
 1
 − 1   − 1  0 1 1 :
R2  , R3  ~
 5   6  0 1 0 : 0
 
0 0 0 : 0

Which is in Echelon form.

⸫ Rank of [A:B] =3=Rank of A=Number of unknowns.

The given systems of equations are consistent and have a unique solution. Hence equation (1) can be
written as

1 2 1  2
0   x  
 1 1    1 
y =
0 1 0   0 
  z   
0 0 0   0 

 x + 2 y + z = 2, y + z = 1, y = 0.
Solving x = 1, y = 0, z = 1.

24
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Example 3: Examine the following equations for consistency and if consistent, find the complete solution.

x + 2 y − z = 3,3x − y + 2z = 1,2x − 2 y + 3z = 2.

Solution: The matrix equation AX=B is given by

1 2 − 1  x  3
3 −1 2   y  = 1 
 (1)
2 −2 3   z  2
1 2 − 1: 3
 A : B = 3 −1 2 : 1 
2 −2 3 : 2
1 2 −1 : 3 
R21 (− 3), R31 (− 2) ~ 0 −7 5 : −8
0 −6 5 : −4
1 2 −1 : 3 
R23 (− 1) ~ 0 −1 0 : −4
0 −6 5 : −4
1 2 −1 : 3 
R32 (− 6) ~ 0 − 1 0 : −4
0 0 5 : 20 
1 2 − 1: 3
1 
R2 (− 1), R3   ~ 0 1 0 : 4 Which is in Echelon form.
5
0 0 1 : 4
⸫ Rank of [A:B] =3=Rank of A=Number of unknowns.

The given systems of equations are consistent and have a unique solution. Hence equation (1) can be
written as

1 2 − 1  x  3
0 1 0   y  = 4

0 0 1   z  4

 x + 2 y − z = 3, y = 4, z = 4.
Solving x = −1, y = 4, z = 4.

Example 4: Prove that the following equations are consistent and solve
2 x + 4 y − z = 9,3x − y + 5z = 5,8x + 2 y + 9 z = 19.

Solution: The matrix equation AX=B is given by

25
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

2 4 − 1  x   9 
3 − 1 5   y  =  5 
(1)

8 2 9   z  19
2 4 − 1 : 9 
 A : B = 3 − 1 5 : 5 
8 2 9 : 19
2 4 −1 : 9 
R21 (− 1), R31 (− 4) ~ 1 −5 6 : −4 
0 − 14 13 : −17
1 −5 6 : −4 
R12 ~ 2 0 − 1 : 9 
0 − 14 13 : −17 
1 −5 6 : −4 
R21 (− 2 ) ~ 0 14 − 13 : 17 
0 − 14 13 : −17 
1 −5 6 : −4
R32 (1) ~ 0 14 − 13 : 17 
0 0 0 : 0 
1 −5 6 : −4 
R2 1 ( )
~ 0
14 
1 − 13 : 17 
14 14
0 0 0 : 0 
which is in Echelon form.⸫ Rank of [A:B] =2=Rank of A < Number of unknowns.

⸫ The given equations are consistent and have infinite many solutions.The equations(1) becomes

1 −5 6  x  − 4 
0 1 − 13   y  = 17 
 14    14
0 0 0   z   0 
(
 x − 5 y + 6 z = −4, y − 13 z = 17 .
14 14
)
Taking z = k (arbitrary value) x = (− 19k + 29 ) / 14, y = (13k + 17 ) / 14, z = k .

Example 5:

Prove that what values of ,  the equations

x + y + z = 6, x + 2 y + 3z = 10, x + 2 y + z = 

Have (i) no solution (ii) a unique solution (iii) infinite many solutions.

26
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Solution: The matrix equation AX=B is given by

1 1 1 x  6 
1 2 3   y  = 10
 (1)
1 2    z    
1 1 1 :6 
 A : B = 1 2 3 : 10
1 2  :  
1 1 1 :6 
R 21 (− 1), R32 (− 1) ~ 0 1 2 : 4 
0 0  − 3:  − 10
Now consider the following cases:

Case I.If   3 ,then Rank of A = Rank of [A:B]=3=Number of unknowns.

Hence in this case the equations are consistent and will have a unique solution.

Case II. If  = 3,  = 10 ,then Rank of A = Rank of [A:B]= 2 < 3 (Number of unknowns).

Hence in this case the equations are consistent and will have infinite many solutions.

Case III. If  = 3,   10 ,then Rank of A = 2,Rank of [A:B]= 3. Therefore ,Rank of A≠ Rank of

[A:B].

Hence in this case the equations are inconsistent and have no solution.

Example 6:

Prove that what values of ,  the equations

x + y + z = 1, x + 2 y + 4 z =  , x + 4 y + 10 z = 2

Have a solution and solve completely in each case.

Solution: The matrix equation AX=B is given by

1 1 1   x   1 
1 2 4   y  =   
     (1)
1 4 10  z  2 
1 1 1 : 1 
 A : B = 1 2 4 :  
1 4 10 : 2 
1 1 1 : 1 
R21 (− 1), R31 (− 1) ~ 0 1 3 :  − 1 
0 3 9 : 2 − 1

27
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1 1 1 :1 
R32 (− 3) ~ 0 1 3 :  − 1 
0 0 0 : 2 − 3 + 2
Deleting the last column of matrix [A:B] clearly rank of A=2,Therefore the equations are consistent if
the rank of the augmented matrix [A:B] is also 2.

 2 − 3 + 2 = 0   = 1,2

Thus the equations have infinite many solutions when  = 1 or  = 2 .

Case I.When  = 1 ,then equation (1)is given by

1 1 1  x  1
0 1 3  y  = 0

0 0 0  z  0

 x + y + z = 1, y + 3z = 0

 If z = k , then x = 2k + 1, y = −3k.

Case II.When  = 2 ,then equation (1) is given by

1 1 1  x  1
0 1 3  y  = 1

0 0 0  z  0

 x + y + z = 1, y + 3z = 1

 If z = k , then x = 2k , y = −3k + 1.

Example 7: Solve the following system of equations :

3x − y + z = 0,−15x + 6 y − 5z = 0,5x − 2 y + 2z = 0.

Solution: The given system of equations is homogeneous and hence it is consistent.

 3 −1 1  x
AX = − 15 6 − 5  y  = 0.

 5 − 2 2   z 
 3 −1 1 
A = − 15 6 − 5

 5 − 2 2 
 3 − 1 1
R21 (5), R31 (− 2) ~  0 1 0
− 1 0 0

28
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

3 − 1 1
R3 (− 1) ~ 0 1 0
1 0 0
1 0 0
R13 ~ 0 1 0
3 − 1 1
1 0 0
R31 (− 3) ~ 0 1 0
0 − 1 1
1 0 0
R32 (1) ~ 0 1 0
0 0 1

Hence Rank of A=3=Number of unknowns.

Hence the system of equations have Trivial solutions which is x = y = z = 0.

Example 8: Solve the following system of equations :

2w + 3x − y − z = 0,4w − 6 x − 2 y + 2 z = 0,−6w + 12 x + 3 y − 4z = 0.

Solution: The given system of equations is homogeneous and hence it is consistent.

 w
 2 3 −1 − 1  
 2    = 0.
x
Here AX =  4 −6 −2
 y (1)
− 6 12 3 − 4  
z

 2 3 −1 − 1
A =  4 −6 −2 2 
− 6 12 3 − 4

2 3 − 1 − 1
R21 (− 2), R31 (3) ~ 0 − 12 0 4 
0 21 0 − 7

2 3 − 1 − 1
 1 1 
R2  − , R3   ~ 0 3 0 − 1
 4 7
0 3 0 − 1

⸫ Rank of A=2< Number of unknowns,(n=4).

The system of equations is consistent and have infinite many solutions.

29
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Here n-rank of A = 4-2=2.

Hence arbitrary values will be given to two unknowns.

 w
2 3 − 1 − 1  
 0 − 1   = 0.
x
Now equation (1) becomes AX = 0 3
  y
0 0 0 0   
z

 2w + 3x − y − z = 0,3x − z = 0

1 1
 If y = k1 , z = k 2 , then x = k 2 , w = k1 .
3 2

Example 9 : Show that the equations :

− 2 x + y + z = a, x − 2 y + z = b, x + y − 2 z = c.

Have no solution unless a + b + c = 0 in which case they have infinitely many solutions. Find the
solutions when a = 1, b = 1, c = −2.

Solution: The matrix equation AX=B is given by

− 2 1 1   x  a 
 1 −2 1   y  = b 
 (1)
 1 1 − 2  z   c 
− 2 1 1 : a

 A : B =  1 − 2 1 : b 
 1 1 − 2 : c 
− 1 − 1 2 : a + b 
R12 (1), R13 ~  1 − 2 1 : b 
 1 1 − 2 : c 
0 0 0 : a + b + c
R13 (1), R23 (− 1) ~ 0 − 3 3 : b − c 

1 1 − 2 :c 
1 1 −2 :c 
 −1  
R13 , R2  −1 :
(c − b) 
 ~ 0 1 
 3   3
0 0 0 : a + b + c 

30
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Deleting the last column of augmented matrix [A:B],clearly the rank of A=2.Therefore ,if the given,
system of equations have a solution then the rank of the augmented matrix must also be 2,which is
possible only if a+b+c=0.

Hence the equations have no solution unless a+b+c=0.

When a+b+c=0, the matrix equation (1) reduces to

1 1 − 2  x   c 
0 1   
− 1  y  =  (c − b ) 
 3
0 0 0   z   0 

 x + y − 2 z = c, y − z = (c − b)
3

 when a = 1, b = 1, c = −2, we have x + y − 2z = −2 and y − z = −1

 Taking z = k, we have y = k -1, x = -2 - k + 1 + 2k i.e. x = k -1, y = k -1, z = k.

Since k is arbitrary,therefore, the given equations have infinite many solutions.

Example 10 : Solve

x + 2 y + 3z = 14,3x + y + 2z = 11,2x + 3 y + z = 11

Solution:

The matrix equation AX=B is given by

1 2 3  x  14
3 1 2  y  = 11
 (1)
2 3 1  z  11
1 23 : 14
A : B = 3 12 : 11
2 31 : 11
1 2 3 : 14 
R21 (− 3), R31 (− 2) ~ 0 − 5 − 7 : − 31

0 − 1 − 5 : − 17 
1 2 3 : 14
R2 (− 1), R3 (− 1) ~ 0 5 7 : 31
0 1 5 : 17 
1 2 3 : 14
R23 ~ 0 1 5 : 17 
0 5 7 : 31

31
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1 2 3 : 14 
R32 (− 5) ~ 0 1 5 : 17 
0 0 − 18 : − 54
1 2 3 : 14
R3 − 1( ~ 0
18 
) 1 5 : 17  whichis in Echelon form.
0 0 1 : 3 

Hence Rank of [A:B]=3=Rank of A =Number of unknowns.

The give equations are consistent and have a unique solution. Hence equation (1) can be
written as follows:
1 2 3  x  14
0 1 5  y  = 17 

0 0 1  z   3 
 x + 2y + 3z = 14
y + 5z = 17
z=3
Solving we get, x=1,y=2,z=3.

Practice Questions:
1 What will be the value of  . so that
( −1)x + (3 + 1)y + 2z = 0
( −1)x + (4 − 2)y + ( + 3)z = 0
2 x + (3 + 1)y + 3( − 1)z = 0.
are consistent. For those value of  . find the values of x,y,z.
Ans:  = 0 , x = y = z = k ;  = 3, y = k1 , z = k 2 then x = -(5k1 + 3k2 )

2 Examine the following equations for consistency and if consistent, find the completesolution:
5x + 3 y + 14z = 4; y + 2z = 1 : x − y + 2z = 0,2x + y + 6z = 2,
Ans: Inconsistent

3 Determine the values of a and b for which the system:

x + y + z = 3, x + 2 y + 2z = 6, x + ay + 3z = b
Have (i) no solution(ii) a unique solution(iii) infinite number of solutions.
Ans (i) a=3,b≠9(ii) a≠3, b any value (iii) a=3,b=9.

4 Determine whether the following equations will have non- trivial solutions, if so solve them;
4x + 2 y + z + 3w = 0;6x + 3 y + 4z + 7w = 0 : 2x + y + w = 0.
Ans:x=(k1+k2)/2, y =k2,z=-k1,w=k1,where k1 and k2 are two are arbitrary constants.

5 Solve the system of equations

x + 3 y − 2z = 0;2x − y + 4z = 0 : x −11y + 14z = 0.
Ans: x=-10/7k, y=8/7k, z=k, where k is arbitrary.

6 Find the value of  . for which the equations

32
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

(11-λ ) x-4 y-7 z = 0

7 x-(λ + 2) y-5 z = 0
10 x- 4 y-(6 + λ )z = 0
possess a non-trivial solution.,For these values of  . ,find the solution also.
Ans :  = 0,1,2
when  = 0, solution is (k, k, k )
when  = 1, solution is (k,-k,2k)
when  = 2, solution is (2k, k,2k )

7 Investigate the values of  and  ,so that the equations

2x + 3 y + 5z = 9,7 x + 3 y − 2z = 8,2x + 3 y + z = 
have (a) no solution, (b) a unique solution and (c) an infinite number of solutions.
Ans : (a ) = 5 ,   9 (b)  5 ,  arbitrary (c) = 5 ,  = 9

8 Verify that the following set of equations have a non-trivial solution:

x + 3 y − 2z = 0,2x − y + 4z = 0, x −11y + 14z = 0
9 Verify that the following system of equations has always a solution:
4x + 3 y − z = 0,3x + 4 y + z = 0, x − y − 2z = 0,5x + y − 4z = 0.

10 Solve the following system of equations:

x1 + x2 + 2 x3 + x4 = 5,2 x1 + 3x2 − x3 − 2 x4 = 2,4 x1 + 5x2 + 3x3 = 7.
Ans: No solution, system inconsistent

1.13 Solution of System of Linear Equations (Gauss-Seidal Method)

This is a modification of Gauss-Jacobi’s method.

Let us consider the system of equations

𝑎11 𝑥1 + 𝑎12 𝑥2 + 𝑎13 𝑥3 = 𝑏1

𝑎21 𝑥1 + 𝑎22 𝑥2 + 𝑎23 𝑥3 = 𝑏2

𝑎31 𝑥1 + 𝑎32 𝑥2 + 𝑎33 𝑥3 = 𝑏3 … … … … … … … … … (1)

Where the diagonal coefficients are not zero and are large compared to other coefficients. Such
a system is called a diagonally dominant system.

The system of equation (1) may be written as

1
𝑥1 = [𝑏 − 𝑎12 𝑥2 − 𝑎13 𝑥3 ]
𝑎11 1

33
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1
𝑥2 = [𝑏 − 𝑎21 𝑥1 − 𝑎23 𝑥3 ]
𝑎22 2

1
𝑥3 = [𝑏 − 𝑎31 𝑥1 − 𝑎32 𝑥2 ] … … … … … . . (2)
𝑎33 3

Let the initial approximate solution be𝑥1 (0) , 𝑥2 (0) , 𝑥3 (0) .

Substituting 𝑥2 (0) , 𝑥3 (0) in the first equation of (2), we get

1
𝑥1 (1) = [𝑏 − 𝑎12 𝑥2 (0) − 𝑎13 𝑥3 (0) ] … … … … . (3)
𝑎11 1

This is taken as the first approximation of x1.

Substituting 𝑥1 (1) for x1 and 𝑥3 (0) for x3 in the second equation of (2), we get

1
𝑥2 (1) = [𝑏2 − 𝑎21 𝑥1 (1) − 𝑎23 𝑥3 (0) ] … … … … … … … (4)
𝑎22

This is taken as the first approximation of x2.

Next, substituting 𝑥1 (1) for x1 and 𝑥2 (1) for x2 in the last equation of (2), we get

1
𝑥3 (1) = [𝑏 − 𝑎31 𝑥1 (1) − 𝑎32 𝑥2 (1) ] … … … … . . (5)
𝑎33 3

This is taken as the first approximation of x3.

The values obtained in (3),(4),(5) constitute the first iterates of the solution.

Proceeding in the same way, we get successive iterates.

The (k+1)th iterates are given by

1
𝑥1 (𝑘+1) = [𝑏1 − 𝑎12 𝑥2 (𝑘) − 𝑎13 𝑥3 (𝑘) ]
𝑎11

1
𝑥2 (𝑘+1) = [𝑏 − 𝑎21 𝑥1 (𝑘+1) − 𝑎23 𝑥3 (𝑘) ]
𝑎22 2

1
𝑥3 (𝑘+1) = [𝑏 − 𝑎31 𝑥1 (𝑘+1) − 𝑎32 𝑥2 (𝑘+1) ] … … … … … (6)
𝑎33 3

34
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

The iteration process is stopped when the desired order of approximation is reached or two
successive iterations are nearly the same. The final values of x1,x2,x3 so obtained constitute an
approximate solution of the system (2).

This method can be generalized to a system of ‘n’ equations ‘n’ unknowns.

The method is known as Gauss-Seidel iteration method.

1. Use Gauss-Seidel iteration method to solve the system 10x+y+z = 12,

2x+10y+z =13, 2x+2y+10z =14
Solution: The given system is diagonally dominant.
Then we can write the equations
1
𝑥= [12 − 𝑦 − 𝑧] … … … (1)
10

1
𝑦= [13 − 2𝑥 − 𝑧] … … … … . (2)
10

1
𝑧= [14 − 2𝑥 − 2𝑦] … … … … . . (3)
10

We start iteration by taking y = 0, z = 0 in equation (1), we get

1 1
𝑥 (1) = [12 − 0 − 0] = [12] = 1.2
10 10
Putting 𝑥 = 𝑥 (1) =1.2, z = 0 in equation (2), we get
1 1 1
𝑦 (1) = [13 − 2𝑥 (1) − 𝑧] = [13 − 2(1.2) − 0] = [10.6] = 1.06
10 10 10
Putting 𝑥 = 𝑥 (1) =1.2, 𝑦 = 𝑦 (1) = 1.06 in equation (3), we get
1 1 1
𝑧 (1) = [14 − 2𝑥 (1) − 2𝑦 (1) ] = [14 − 2(1.2) − 2(1.06)] = [9.48] = 0.948
10 10 10
Now taking 𝑦 (1) , 𝑧 (1) as the initial values in equation (1), we get
1 1 1
𝑥 (2) = 10 [12 − 𝑦 (1) − 𝑧 (1) ] = 10 [12 − 1.06 − 0.948] = 10 [9.992] = 0.9992
Taking 𝑥 = 𝑥 (2) and 𝑧 = 𝑧 (1) in equation (2), we get
1 1 1
𝑦 (2) = [13 − 2𝑥 (2) − 𝑧 (1) ] = [13 − 2(0.999) − 0.948] = [10.054]=1.0054
10 10 10
Next taking 𝑥 = 𝑥 (2) =0.999, 𝑦 = 𝑦 (2) = 1.005 in equation (3), we get
1 1 1
𝑧 (2) = 10 [14 − 2𝑥 (2) − 2𝑦 (2) ] = 10 [14 − 2(0.999) − 2(1.005)] =10 [9.992]= 0.999

Again taking 𝑥 (2) ,𝑦 (2) ,𝑧 (2) as the initial values, we get

1 1 1
𝑥 (3) = 10 [12 − 𝑦 (2) − 𝑧 (2) ] = 10 [12 − 1.005 − 0.999] = 10 [9.996]= 0.9996=1.00
1 1 1
𝑦 (3) = 10 [13 − 2𝑥 (3) − 𝑧 (2) ] = 10 [13 − 2(1) − 0.999]= 10 [10.001]= 1.0001=1
1 1 1
𝑧 (3) = 10 [14 − 2𝑥 (3) − 2𝑦 (3) ] = 10 [14 − 2(1) − 2(1)]= 10 [10] = 1
Again taking 𝑥 (3) ,𝑦 (3) ,𝑧 (3) as the initial values, we get

35
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1 1 1
𝑥 (4) = 10 [12 − 𝑦 (3) − 𝑧 (3) ] = 10 [12 − 1 − 1] = 10 [10] = 1
1 1 1
𝑦 (4) = 10 [13 − 2𝑥 (4) − 𝑧 (3) ] = 10 [13 − 2(1) − 1] = 10 [10] = 1
1 1 1
𝑧 (4) = 10 [14 − 2𝑥 (4) − 2𝑦 (4) ] = 10 [14 − 2(1) − 2(1)] = 10 [10] = 1

⸫ The solution of given equations are x = 1, y = 1, z = 1.

2. Solve the following system of equations by using Gauss-Seidel method correct to three
decimal places.
8x-3y+2z = 20,4x+11y-z = 33, 6x+3y+12z = 35.
Solution: Consider the system of equations
8x-3y+2z = 20
4x+11y-z = 33
6x+3y+12z = 35 ………… (1)
Since the diagonal elements are dominant in the coefficient matrix of equation (1)
i.e: |8| > |−3| + |2|
|11| > |4| + |−1|
|12| > |6| + |3|

Convergence condition is satisfied, we apply iterative method for the given system of (1), then
we can write x,y,z as follows.
1
𝑥 = [20 + 3𝑦 − 2𝑧] … … … … … . (2)
8

1
𝑦= [33 − 4𝑥 + 𝑧] … … … … … . (3)
11

1
𝑧=[35 − 6𝑥 − 3𝑦] … … … … . . (4)
12
Iteration :( 1) Puttingy=0,z= 0 in R.H.S of equation (2), we get

1 1 1
𝑥1 = 8 [20 + 3𝑦 − 2𝑧] = 8 [20 + 0 − 0] = 8 [20] = 2.5
Putting𝑥 = 𝑥1 = 2.5, z= 0 in R.H.S of equation (3), we get
1 1 1
𝑦1 = 11 [33 − 4𝑥1 + 𝑧] = 11 [33 − 4(2.5) + 0] = 11 [23] = 2.091
Putting𝑥 = 𝑥1 = 2.5, 𝑦 = 𝑦1 = 2.091 in R.H.S of equation (4), we get
1 1 1
𝑧1 = 12 [35 − 6𝑥1 − 3𝑦1 ] = 12 [35 − 6(2.5) − 3(2.091)] = 12 [13.727] = 1.144

Iteration :( 2) For second approximation

1 1 1
𝑥2 = 8 [20 + 3𝑦1 − 2𝑧1 ] = 8 [20 + 3(2.091) − 2(1.144)] = 8 [23.985] = 2.998

1 1 1
𝑦2 = 11 [33 − 4𝑥2 + 𝑧1 ] =11 [33 − 4(2.998) + 1.144] = 11 [22.152] = 2.014

36
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

1 1 1
𝑧2 = 12 [35 − 6𝑥2 − 3𝑦2 ] = 12 [35 − 6(2.998) − 3(2.014)] = 12 [10.97] = 0.914

Iteration :( 3) For third approximation

1 1 1
𝑥3 = 8 [20 + 3𝑦2 − 2𝑧2 ] =8 [20 + 3(2.014) − 2(0.914)] = 8 [24.214] = 3.027
1 1 1
𝑦3 = 11 [33 − 4𝑥3 + 𝑧2 ] = 11 [33 − 4(3.027) + 0.914] = 11 [21.806] = 1.982
1 1 1
𝑧3 = 12 [35 − 6𝑥3 − 3𝑦3 ] = 12 [35 − 6(3.027) − 3(1.982)] = 12 [10.892] = 0.908

Iteration :( 4) For fourth approximation

1 1 1
𝑥4 = 8 [20 + 3𝑦3 − 2𝑧3 ] = 8 [20 + 3(1.982) − 2(0.908)] = 8 [24.13] = 3.016
1 1 1
𝑦4 = 11 [33 − 4𝑥4 + 𝑧3 ] = 11 [33 − 4(3.016) + 0.908] = 11 [21.844] = 1.986
1 1 1
𝑧4 = 12 [35 − 6𝑥4 − 3𝑦4 ] = 12 [35 − 6(3.016) − 3(1.986)] = 12 [10.946] = 0.912

Iteration :( 5) For fifth approximation

1 1 1
𝑥5 = 8 [20 + 3𝑦4 − 2𝑧4 ] = 8 [20 + 3(1.986) − 2(0.912)] = 8 [24.134] = 3.017
1 1 1
𝑦5 = 11 [33 − 4𝑥5 + 𝑧4 ] = 11 [33 − 4(3.017) + 0.912] = 11 [21.844] = 1.986
1 1 1
𝑧5 = 12 [35 − 6𝑥5 − 3𝑦5 ] = 12 [35 − 6(3.017) − 3(1.986)] = 12 [10.94] = 0.912

Iteration :( 6) For sixth approximation

1 1 1
𝑥6 = 8 [20 + 3𝑦5 − 2𝑧5 ] = 8 [20 + 3(1.986) − 2(0.912)] = 8 [24.134] = 3.017
1 1 1
𝑦6 = 11 [33 − 4𝑥6 + 𝑧5 ] = 11 [33 − 4(3.017) + 0.912] = 11 [21.844] = 1.986
1 1 1
𝑧6 = 12 [35 − 6𝑥6 − 3𝑦6 ] = 12 [35 − 6(3.017) − 3(1.986)] = 12 [10.94] = 0.912

We observe that from 5th and 6th iterations are same.

⸫ The solution of the equations are x = 3.017, y = 1.986, z = 0.912

Practice Question

Solve the following system of equations by using Gauss-Seidel method correct to three decimal
places.

(i) x+y+54z = 110, 27x+6y-z = 85,6x+15y+2z = 72

37
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

(ii) 8x+y+z = 8, 2x+4y+z = 4, x+3y+6z = 5

1.14 Linear Dependence and Independence of vectors

Let S = {X1, X2….Xn}be a set of vectors (row matrices or column matrices).

(A) X1,X2,……..Xnare said to be linearly dependent (L.D.) if

(i) X1,X2,……..Xn all are of the same order,
(ii) There exist n-scalars (or numbers) a1,a2,………,an not all of them zero (some of them may be
zero), such that
a1X1+ a2X2+……+ anXn= 0. …………………….(1)

If a1≠ 0,then transposing a1X1 to the other side and dividing by (-a1), the relation (1)may be written as

X = b2X2+ b3X3+…….+bnXn. …………………………………..(2)

The relation (2) shows that the vector X1 is a linear combination of the vectors X2,X3,….Xn.

(B) The vectors X1,X2,…………..Xn are said to be linearly independent (L.I) if every relation of the
form
a1 X1+ a2X2+……+ an Xn= 0.


a1=0,a2=0, ………..an=0 i.e., all the n scalars a1, a2……………an vanish.

Example 1: Prove that the four vectors X 1 = (1,2,3), X 2 = (1,0,0), X 3 = (0,1,0), X 4 = (0,0,1) are
linearly dependent. Find the relation between them.

Solution: Consider the vector (or matrix equation)

a1 X 1 + a2 X 2 + a3 X 3 + a4 X 4 = 0
 a1 (1,2,3) + a2 (1,0,0) + a3 (0,1,0) + a4 (0,0,1) = (0,0,0) (1)

a1 + a2 = 0,2a1 + a3 = 0,3a1 + a4 = 0 (2)

Taking a1 =   0, we obtain from equations(2),

 a2 = − , a3 = −2 , a4 = −3

Hence the given vectors are linearly dependent. Putting the values of a1 , a2 , a3 , a4 in (1), we get

X 1 − X 2 − 2X 3 − 3X 4 = 0

X 1 − X 2 − 2 X 3 − 3 X 4 = 0 which is the required solution.

Example 2: Examine for linear dependence X 1 = (2,−1,3,0), X 2 = (1,2,−1,−5), X 3 = (1,3,−2,−7).

Solution: Consider the vector equation

38
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

a1 X 1 + a2 X 2 + a3 X 3 = 0
 a1 (2,−1,3,0) + a2 (1,2,−1,−5) + a3 (1,3,−2,−7 ) = (0,0,0,0) (1)
 2a1 + a2 + a3 = 0,-a1 + 2a2 + 3a3 = 0,3a1 − a2 − 2a3 = 0,0a1 − 5a2 − 7a3 = 0 (2)
The system (2) of equations is homogeneous and so it is consistent. We have
2 1 3 0 
- 1 2   a1   
 3    0 
a =
 3 - 1 - 2   2  0 
   a3   
 0 - 5 - 7  0 
2 1 3
- 1 2 3 
Coefficient matrix A = 
3 - 1 - 2
 
0 - 5 - 7

- 1 2 3
2 1 3 
R 12 ~
 3 -1 - 2
 
0 -5 - 7
- 1 2 3
0 5 9 
R 21 (2 ), R31 (3) ~ 
0 5 7
 
0 - 7
-5
- 1 2 3 
0 5 9 
R 43 (1), R32 (− 1) ~  . Rank of A = 3 = number of unknowns.
 0 0 - 2
 
0 0 0 
Hence the system (2) of equations has only trivial solution i.e. a1 = 0, a2 = 0, a3 = 0.

Hence the given vectors are linearly independent.

Example 3: Find the value of  for which the vectors (1,−2,  ), (2,−1,5) and (3,−5,7 ) are linearly
dependent.

Solution: The given vectors will be linearly dependent if one vector can be expressed as a linear
combination of the remaining vectors. Let

(1,−2,  ) = a1 (2,−1,5) + a2 (3,−5,7 ) (1)

 1 = 2a1 + 3a2 (2)
 −2 = −a1 − 5a2 (3)
  = 5a1 + 7a2 (4)
1 3
Solving (2) and (3); a1 = − , a 2 = .
7 7

39
 KIET GROUP OF INSTITUTIONS, DELHI NCR, GHAZIABAD

Putting for a1 and a2 in (4), we get

5 5
 = − + 3   = .
7 14

Example 4: Test the following vectors (1,2,3),(3,-2,1),(1,-6,-5) for linear dependence:

Solution:

1 3 1

Coefficient matrix, A = 2 − 2 − 6
 
3 1 − 5

1 3 1

R21 (− 2), R31 (− 3)0 − 8 − 8
0 − 8 − 8
1 3 1

R32 (− 1), 0 − 8 − 8  Rank of A = 2( 3, number of vectors)
0 0 0 
Hence the given vectors are linearly dependent.

Practice Question

Test the following system of vectors for linear dependence. If dependent, establish the relation
between then:

1) X 1 = (1,1,−1), X 2 = (2,−3,5), X 3 = (−2,1,4).

Ans: Linearly independent.

2) X 1 = (1,1,0,0), X 2 = (0,1,−1,0), X 3 = (0,0,0,3).

Ans: Linearly independent.

3) X 1 = (1,0,2,1), X 2 = (3,1,2,1), X 3 = (4,6,2,−4), X 4 = (−6,0,−3,−4).

Ans:Dependent,2X1-6X2+X3-2X4=0.

4) X 1 = (1,1,−1,1), X 2 = (1,−1,2,−1), X 3 = (3,1,0,1).

Ans:Dependent,2X1+X2-X3=0

5) X 1 = (1,3,2), X 2 = (1,−7 − 8), X 3 = (2,1,−1).

Ans:Dependent,3X1+X2-2X3=0.

40