2.

2 - Solving Ax=0 And Ax=b

Vocab : Cofficient Mtrix,Augmented Matrix,EchelonForm,Row Reduced Form ,Rank,Pivot

Variable,free Variable

Ex 1 - Consider a System of Linear Equation


x1 − 2x2 + x3 = 0 


2x2 − 8x3 = 8 (1)


−4x1 + 5x2 + 9x3 = −9

1
solve (1)

2
 Ex 2 - A system of linear equations is a list of linear equations with the same unknowns. In
particular, a system of 2 linear equations in 2 unknowns x1 , x2 can be put in the standard form

3
 )
a11 x1 + a12 x2 = b1
(2)
a21 x1 + a22 x2 = b2
where aij , bi are constant.and we can rewrite system (2) as :

 
a11 x1 + a12 x2 = b1
(3)
a21 x1 + a22 x2 = b2
again we can rewrite system (3) to as
     
a11 a12 x1 b
= 1 (4)
a21 a22 x2 b2
again we can rewrite system (4)(without using unknown,for the simplicity) to as

 
a11 a12 b1
(5)
a21 a22 b2
   
a11 a12 b1 a a
A= , C = 11 12
a21 a22 b2 a21 a22
where A is called Augmented Matrix and C is called Cofficient Matrix of the system

R1 → a 1 R1 1 aa12 b1
   
a11 a12 b1 11 a11
−−−−−−→ 11
a21 a22 b2 a21 a22 b2
 a12 b1 
R2 →R2 −a21 R1 1 a11 a11
−−−−−−−−−→ (6)
0 a22 − a21a11a12 b2 − aa2111b1
(6) can be written as

a12 b1
)
x1 + x
a11 2
= a11
a21 a12 a21 b1
(7)
0x1 + (a22 − a11
)x2 = b2 − a11
a21 b1
a21 a12 b2 −a11 a12 b1
case 1 : if a22 − a11
6= 0 then x2 = a a12
a22 − 21
, x1 ca be calculated from x1 + x
a11 2
= a11
a11
Conclusion- unique solution of (2)
Case 2 : if a22 − a21a11a12 = 0 and b2 − aa2111b1 = 0 then second equation of (7) becomes 0x1 +0x2 = 0
Conclution - infinite solution of (2)
Case 3 : if a22 − a21a11a12 = 0 and b2 − aa2111b1 6= 0 then second equation of (6)becomes 0x1 + 0x2 =
b2 − aa2111b1 i.e. 0 = b2 − aa2111b1 which is not true
Conclution - solution does not exit of (2).

4
 Elementary Row Operations

Echelon Matrices (or in echelon form) U and Row Reduced Form R

Echelon Matrices U
A Matrix U is called an echelon matrix or is said to be in echelon form , if the following two
conditions hold :
(1) All zero rows,if any, are at the bottem of the matrix.
(2) Each leading nonzero entry in a row is to the right of the leading nonzero entry in the
preceding row.

Row Reduced Form R

A Matrix is said to be in row reduced form R if it is an echelon matrix and if satisfies the
following additional two properties:
(3)Each pivot(leading nonzero entry) is equal to 1.
(4) Each pivot is the only nonzero entry in its column.

EX 3 The following is an echelon matrix whose pivots have been circled

5
 NOTE 1 - The major difference between an echelon matrix in row reduced form is that in
an echelon matrix there must be zeros below the pivots [properties(1)and (2)] but in a matrix
in row reduced form , each pivot must also equal 1 [property (3)] and there must also be zeros
above the pivots [properties(4)].

Ex-4 The following are echelon matrices whose pivots have been circled

The Third matrix is also an example of a matrix in row reduced form. the second matrix is not
in row reduced form ,since it does not satisfy property(4),taht is,there is a nonzero entry above
the second pivot in the third column.The first matrix is not in row reduced form, because it
satisfies neither property (3) nor property (4); that is, some pivots are not equal to 1 and there
are nonzero entries above the pivots.

Ex-5 The entries of a 5 by 8 echelon matrix U and its reduced form R

6
 Pivot Variable and Free Variable
Pivot Variable: Pivot Variable are those variable that correspond to columns with pivots.
Free Variable : Free Variable are those variable that correspond to columns without pivots.

Note : If Ax = 0 has more unknowns than equations (n > m), it has at least one special
solution: There are more solutions than the trivial x = 0.
Note : xcomplete = xparticlar + xnullspace
Note : if there are n column in a matrix A and there are r pivots then there are r pivot variables
and n − r free variable.and this important number r is called Rank of a Matrix.

Rank of a Matrix = The rank of a matrix A, written rank(A), is equal to the maximum
number of linearly independent columns of A
= number of pivot column in the echelon form of a matrix A
=maximum number of linearly independent rows of A
= dimension of the column space of A
= dimension of the row space of A.

Note : Let A be an n-square matrix. then A is invertible if and only if rank(A) = n

Ex 6: Find Rank of A  
1 2 3 5
A= 0 0 2 2 
0 0 0 0

Sol. Since Echelon form of A is itself A.and 1st and 3rd column are pivot column.
So Rank of A is 2.

Method for solving System of linear equation

Method-1

7
Ex 7 - Consider a System of linear equation


1x1 + 2x2 + 3x3 + 5x4 = b1 


2x1 + 4x2 + 8x3 12x4 = b2 (8)


3x + 6x + 7x + 13x = b 
1 2 3 4 3

Sol.
Step 1: Reduce Ax = b to U x = c
i.e. Reduce Augmented Matrix [A b] to Augmented Matrix [U c]

   
1 2 3 5 b 1 1 2 3 5 b 1
  R2 →R2 −2R1 ,R3 →R3 −3R1
A b =  2 4 8 12 b2  −−−−−−−−−−−−−−−→  0 0 2 2 b2 − 2b1 
3 6 7 13 b3 0 0 −2 −2 b3 − 3b1
 
1 2 3 5 b1
R →R +R2
−−3−−−3−−→  0 0 2 2 b2 − 2b1  (9)
0 0 0 0 b3 + b2 − 5b1
 
= U c

 

 (9)means 

1x + 2x + 3x + 5x = b
 
1 2 3 4 1

 

 
 0x1 + 0x2 + 2x3 + 2x4 = b2 − 2b1

 


0x1 + 0x2 + 0x3 + 0x4 = b3 + b2 − 5b1
third equation hold only if b3 + b2 − 5b1 = 0

 


 

it means if b + b − 5b = 0 then system of equation has infinite solution.
 
3 2 1

 

 
if b3 + b2 − 5b1 6= 0 then system of equation has no solution.
 

Here    
1 2 3 5 b1
U =  0 0 2 2  , C =  b2 − 2b1 
0 0 0 0 b3 + b2 − 5b1
Step 2 :
Find Special Solution : U x = 0
Take particularly b1 = 0,b2 = 6,b3 = −6

 
  1 2 3 5 0
U 0 = 0 0 2 2 0  (10)
0 0 0 0 0

Here x2 and x4 are free variables

Let x2 = a,x4 = b where a,b belongs to Set of Real Number

8
Now we can rewrite (10) as
1x1 + 2x2 + 3x3 + 5x4 = 0—–(*)
0x1 + 0x2 + 2x3 + 2x4 = 0—–(**)
now put the value of x2 in (**)
2x3 + 2b = 0
i.e. x3 = −b
now put the value of x3 in (*)
x1 + 2a − 3b + 5b = 0
i.e. x1 + 2a + 2b = 0
i.e. x1 = −2a − 2b  
−2a − 2b
 a 
Special Solution xn = 
 −b 


b
   
−2 −2
1 0
xn = a 
 0  + b  1  where a, b belongs to set of real number
  

0 −1
Step 3 :
Find Particular Solution xp , U xp = c and put all free variables= 0

So put x2 = a = 0, x4 = b = 0

 
  1 2 3 5 0
U c = 0 0 2 2 6  (11)
0 0 0 0 0

(11) Can be rewritten as

1x1 + 2x2 + 3x3 + 5x4 = 0—–(*)
0x1 + 0x2 + 2x3 + 2x4 = 6—–(**)
Now put b = 0 in (**)
2x3 + 0 = 6
x3 = 3
Now a = 0,x3 = 3,b = 0 in (*)
x1 0 + 9 + 0 = 0
x1 = −9
 
−9
0
xp = 
3


9
 Step 4 :        
−2 −2 −9 −2a − 2b − 9
1
 + b 0  +  0  =  a
     
Complete Solution x = xn + xp = a  0  1   3   −b + 3 


0 −1 0 b
     
−9 −2a −2b
 0   a   0b 
= 3  +  0a  +  b 
    

0 0a −b
     
−9 −2 −2
0
 + a 1  + b 0 
   
=3 0 1
0 0 −1
where a, b belongs to Set of real numbers

Method-2
Ex - Consider a System of linear equation


1x1 + 2x2 + 3x3 + 5x4 = 0 


2x1 + 4x2 + 8x3 12x4 = 6 (12)


3x1 + 6x2 + 7x3 + 13x4 = −6
Step 1: Reduce Ax = b to U x = c
i.e. Reduce Augmented Matrix [A b] to Augmented Matrix [U c]

   
1 2 3 5 0 1 2 3 5 0
  R2 →R2 −2R1 ,R3 →R3 −3R1
A b =  2 4 8 12 6  −−−−−−−−−−−−−−−→  0 0 2 2 6 
3 6 7 13 −6 0 0 −2 −2 −6
 
1 2 3 5 0
R →R +R2
−−3−−−3−−→  0 0 2 2 6  (13)
0 0 0 0 0
 
= U c

 

 (13)means 

1x1 + 2x2 + 3x3 + 5x4 = 0
 

 0x1 + 0x2 + 2x3 + 2x4 = 6 

0x1 + 0x2 + 0x3 + 0x4 = 0
 

Here    
1 2 3 5 0
U = 0 0 2 2 , c = 6
  
0 0 0 0 0

10
Step 2:
Here x2 and x4 are free variables
Let x2 = a,x4 = b where a,b belongs to Set of Real Number
Now we can rewrite (0.11) as
1x1 + 2x2 + 3x3 + 5x4 = 0—–(*)
0x1 + 0x2 + 2x3 + 2x4 = 6—–(**)
now put the value of x2 in (**)
2x3 + 2b = 6
i.e. x3 = 3 − b
now put the value of x3 in (*)
x1 + 2a + 3(3 − b) + 5b = 0
i.e. x1 + 2a + 9 + 2b = 0
i.e. x1 = −9 − 2a − 2b      
−2 −2 −9
1 0 0
Complete Solution x = xn + xp = a   0  + b 1  +  3 
    

0 −1 0
       
−2a − 2b − 9 −9 −2a −2b
 a  =  0  +  a  +  0b 
      
= −b + 3   3   0a   b 
b 0 0a −b
     
−9 −2 −2
0 1 0
= 3  + a 0  + b 1 
    

0 0 −1
where a, b belongs to Set of real numbers

Exercise 2.2.1 : find the value of c that makes it possible to solve Ax = b, and solve it:

u + v + 2w = 2

2u + 3v − w = 5

3u + 4u + w = c
     
1 1 2 2 1 1 2 2 1 1 2 2
R2 →R2 −2R1 R →R −R2
Solution Aug matrix =  2 3 −1 5  −− −−−−−→  0 1 −5 1  −−2−−−3−−→  0 1 −5 1 
3 4 1 c 0 1 −5 c 0 0 0 c−7
Solution Exit only if c − 7 = 0 so assume w = k ∈ R

v − 5w = 1

11
 v − 5k = 1

v = 1 + 5k

u + v + 2w = 2

u + (1 + 5k) + 2k = 2

u = 1 − 7k

       
u 1 − 7k 1 −7
 v  = 1 + 5k  = 1 + k  5  where k ∈ R
w k 0 1
Exercise 2.2.4 Write the complete solution x = xp + xn to these systems ,(as in equation
(4))
   
  u     u  
1 2 2   1 1 2 2   1
v = , v = ,
2 4 5 4 2 4 4 4
w w
   
1 2 2 1 R2 →R2 −2R1 1 2 2 1
Solution (1) Aug matrix = −−−−−−−→
2 4 5 4 0 0 1 2
Since v is free variable so take v = k , k ∈ R

w=2

u + 2v + 2w = 1

u + 2k + 4 = 1

u = −3 − 2k

       
u −3 − 2k −3 −2
v  =  k  =  0  + k  1 , where k ∈ R
w 2 2 0

   
1 2 2 1 R2 →R2 −2R1 1 2 2 1
(2) Aug matrix = −−−−−−−→
2 4 4 4 0 0 0 2
i.e.
u + 2v + 2w = 1

0u + 0v + 0w = 2

i.e. 0 = 2 which is not true.

So there is no solution.

12
Exercise 2.2.5 Reduce A and B to echelon form, to find their ranks, which variables are
free ?   
1 2 0 1 1 2 3
A = 0 1 1 0 , B = 4 5 6 find the special solutions to Ax = 0 and Bx = 0. find all
1 2 0 1 7 8 9
solutions.
   
1 2 0 1 1 2 0 1
R →R −R1
Solution:(1) A = 0 1 1 0 −−3−−−3−−→ 0 1 1 0 = U. Since first two U are L.I. So
1 2 0 1 0 0 0 0
ρ(A) = 2.
Now for solving Ax = 0.   
1 2 0 1 0 1 2 0 1 0
R →R −R1
Aug. matrix = [A|0] =  0 1 1 0 0  −−3−−−3−−→  0 1 1 0 0 
1 2 0 1 0 0 0 0 0 0
Since x3 and x4 are free variable; so assume x3 = k1 , x4 = k2 , where k1 , k2 ∈ R
x2 + x3 = 0 ⇒ x2 + k = 0 ⇒ x2 = −k

x1 + 2x2 + x4 = 0

x1 − 2k1 + k2 = 0

x1 = 2k1 − k2
       
x1 2k1 − k2 2 −1
x2   −k1  −1 0
 =
x3   k1  = k1  1  + k2  0 
    

x4 k2 0 1
where k1 , k2 ∈ R.
This is general solution.    
2 −1
−1 0
Hence special solutions are 
 1  and  0  .
  

  0 1   
1 2 3 1 2 3 1 2 3
R2 →R2 −4R1 R3 →R3 −2R2
(2) B = 4 5 6 −− −−−−−→ 0 −3 −6  −− −−−−−→ 0 −3 −6 = U
R3 →R3 −7R1
7 8 9 0 −6 −12 0 0 0
Since U has two pivot columns, so ρ(B) = 2.
for solving Bx = 0      
1 2 3 0 1 2 3 0 1 2 3 0
R2 →R2 −4R1 R3 →R3 −2R2
Aug. matrix = [B|0] =  4 5 6 0  −− −−−−−→  0 −3 −6 0  −− −−−−−→  0 −3 −6 0 
R3 →R3 −7R1
7 8 9 0 0 −6 −12 0 0 0 0 0
Since x3 is free variable.
So x3 = k, k ∈ R.

13
 −3x2 − 6x3 = 0

x2 = −2k

x1 + 2x2 + 3x3 = 0

x1 − 4k + 3k = 0

x1 − k = 0 ⇒ x1 = k.
       
x1 k 1 1
General solution x2 = −2k = k −2 , k ∈ R. Special solution is −2.
      
x3 k 1 1

14