SIM1001 Basic Mathematics

SIX1013 Fundamental of Advanced Mathematics

Part II LINEAR ALBEGRA

Chapter 1 Linear Systems and Gaussian Elimination

1.1 Linear Systems and their solution

Definition 1.1.
A linear equation in n variables x1, x2,…, xn has the form
a1x1 + a2x2 + … + anxn = b
where a1, a2, a3, … , an and b are real constants. An ai, , 1 ≤ i ≤ n, is called a coefficient , and the n
variables in the linear equation are also called the unknowns.

Example 1.1.
1. The equations 4x − 2y = 1, x − 4y + 7z = 5, x1 + x2 − 2x3 + x4 + 0.5x5 = 0 and
x1 + x2 + … + xn = 1 are linear equations.

2. The equations xy = 1, 3x2 + 4y2 = 5, sin x + cos y = 1, and x − ey = 0 are non-linear.

Definition 1.2. Given n real numbers s1, s2, s3, … , sn. we say that x1 = s1, x2 = s2, … , xn = sn is a
solution of a linear equation a1x1 + a2x2 + … + anxn = b if the equation is satisfied when we
substitute the values into the equation accordingly. The set of all solutions of the equations is
called the solution set of the equation and an expression that gives us all these solutions is called
the general solution of the equation.

Example 1.2.
1. Consider the linear equation 4x − 2y = 1.
x = 0, y = −1/2 is a solution of 4x − 2y = 1. x = 1/4, y = 0; and x = 1, y = 3/2 are also
solutions of 4x − 2y = 1. The linear equation 4x − 2y = 1 has infinitely many other solutions.
1
x = t, y = 2t − , t ∈ R , (t is an arbitrary parameter) is called the general solution of
2
the linear equation.
⎧ 1 ⎫
The solution set of the linear equation is ⎨(x, y)| x = t, y = 2t − , t ∈ R ⎬ or
⎩ 2 ⎭
⎧⎛ ⎞ ⎫
⎪⎜ t ⎟ ⎪
⎨⎜ 1 ⎟, t ∈ R⎬ .
⎪⎜ 2t − ⎟ ⎪
⎩⎝ 2 ⎠ ⎭
s 1
We also can write the general solution as x = + , y = s, s ∈ R (s is an arbitrary
2 4
parameter).

2. The general solution for the linear equation x − 4y + 7z = 5 is by x = 5 + 4s − 7t, y = s,

z = t, s, t ∈ R (s and t are arbitrary parameters).
1
Definition 1.3. A finite set of m linear equations in n unknowns x1, x2,…, xn is called a system of
linear equations (or linear system):
a11x1 + a12x2 + … + a1nxn = b1
a21x1 + a22x2 + … + a2nxn = b2
!
am1x1 + am2x2 + … + amnxn = bn
where aij and bi, 1 ≤ i ≤ m, 1 ≤ j ≤ n are real constants. An aij is called a coefficient , bi is called a
constant and xj is called an unknown (or variable).

Given n real numbers s1, s2, s3, … , sn. x1 = s1, x2 = s2, … , xn = sn is a solution of the system if it is
a solution of every equation in the system. The set of all solutions of the system is called the
solution set of the system and an expression that gives us all these solutions is called the general
solution of the system.

Example 1.3. Consider the system of linear equations: x − 4y + 7z = 5

2x + 15z = 10.
x = 5, y = 0 , z = 0 is a solution of the system because it satisfies both equations, but x = 9, y = 1 ,
z = 0 is not a solution of the system because it does not satisfy 2x + 15z = 10.

Remark 1.1:
NOT ALL system of linear equations have solutions. For example: the system of linear equations
x − 2y + 3z = 1
−2x + 4y − 6z = 5
has no solution.

Definition 1.4. A system of linear equations that has at least one solution is called consistent. A
system that has no solution is said to be inconsistent.

Remark 1.2:
Every system of linear equations has either no solution, exactly one solution or infinitely many
solutions.

Definition 1.5.
A system of m linear equations in n unknowns x1, x2,…, xn:
a11x1 + a12x2 + … + a1nxn = b1
a21x1 + a22x2 + … + a2nxn = b2
!
am1x1 + am2x2 + … + amnxn = bn
is equivalent to the matrix equation AX = B
⎡ a11 a12 a13 ! a1n ⎤
⎢a a 22 a 23 ! a 2 n ⎥⎥
⎢ 21
where A = is the coefficient matrix, and
⎢ " " " " ⎥
⎢ ⎥
⎣a m1 a m 2 a m3 ! a mn ⎦
⎡ x1 ⎤ ⎡ b1 ⎤
⎢x ⎥ ⎢b ⎥
2 ⎥ and
X = ⎢ B = ⎢ 2 ⎥ are the column vectors of the unknowns and of the constants, respectively.
⎢!⎥ ⎢!⎥
⎢ ⎥ ⎢ ⎥
⎣ xn ⎦ ⎣bn ⎦
If B = 0 (that is, bi = 0 for all i), the system is said to be homogeneous.
If B ≠ 0 (that is, there exists a bi ≠ 0), the system is said to be non-homogeneous.
2
This system of m linear equations can be represented by a rectangular array numbers
⎡ a11 a12 a13 " a1n b1 ⎤
⎢a a 22 a 23 " a 2 n b2 ⎥⎥
[ A | B] = ⎢ 21
⎢ ! ! ! ! !⎥
⎢ ⎥
⎣a m1 a m 2 a m3 " a mn bm ⎦
This array is called the augmented matrix of the system.

⎡ − 1 1 2 2⎤
Example 1.4: The array ⎢ 3 − 1 1 6⎥ is the augmented matrix of the system of linear
⎢ ⎥
⎢⎣− 1 3 4 4⎥⎦
equations −x1 + x2 + 2x3 = 2
3x1 − x2 + x3 = 6
−x1 + 3x2 + 4x3 = 4.

1.2 Elementary row operations and Row Echelon Forms

Definition 1.5: Elementary row operations

A matrix A is said to be row equivalent to a matrix B if B can be obtained from A by a finite
sequence of the following operations called elementary row operations:

[E1]: Interchange the ith row and the jth row: Ri ↔ R j .

[E2]: Multiplication of a row by a nonzero constant λ: Ri → λ Ri , λ ∈ R \{0} .

[E3]: Addition of a constant multiple of one row to another: Ri → Ri + β R j , β ∈ R .

Example 1.5:
⎡ − 1 1 2 2⎤ R1↔R3
⎡ −1 3 4 4 ⎤
[E1]: ⎢ 3 − 1 1 6⎥
⎢ ⎥
⎯ ⎯⎯→ ⎢⎢ 3 −1 1 6 ⎥⎥ .
⎢⎣− 1 3 4 4⎥⎦ ⎢ −1 1 2 2 ⎥
⎣ ⎦

R3 →−3R3 ⎡ −1 1
⎡ − 1 1 2 2⎤ 2 2 ⎤
⎢ ⎥
[E2]: ⎢ 3 − 1 1 6⎥
⎢ ⎥
⎯ ⎯⎯⎯
→ ⎢ 3 −1 1 6 ⎥.
⎢⎣− 1 3 4 4⎥⎦ ⎢ 3 −9 −12 −12 ⎥
⎣ ⎦

⎡ − 1 1 2 2⎤ R2 →R2 +3R3
⎡ −1 1 2 2 ⎤
[E3]: ⎢ 3 − 1 1 6⎥
⎢ ⎥
⎯ ⎯⎯⎯→ ⎢⎢ 0 8 13 18 ⎥⎥ .
⎢⎣− 1 3 4 4⎥⎦ ⎢ −1 3 4 4 ⎥
⎣ ⎦

Theorem 1.1: If augmented matrices of two systems of linear equations are row equivalent, then
the two systems have the same set of solutions.
3
Definition 1.6: Row-Echelon Form
A matrix is said to be row-echelon form if it has the following properties:
1. If there are any zero rows (a zero row is a row with all entries are zeros), then they are grouped
together at the bottom of the matrix.

2. In any two successive non-zero rows (a non-zero row is a row consists at least one non-zero
entry), the first non-zero number in the lower row occurs farther to the right than the first non-
zero number in the upper row. (The first non-zero number in a row is called the leading entry
of the row.)

A matrix is said to be reduced row-echelon form if it is in row-echelon form and has the following
additional properties:

3. The leading entry of every nonzero row is 1.

4. Each column that contains a leading entry has zeros everywhere else.

He

Remark 1.3:
1. In some text books, the row-echelon form is required to have the additional property 3.

2. Every matrix can be reduced to a row-echelon matrix (a matrix in row-echelon form) by

some elementary row operations.

3. Every matrix has a unique reduced row-echelon form but it can have many different row-
echelon forms.

4. The zero matrix 0 (the matrix with all entries are zeros) is a row echelon matrix and also row
reduced echelon matrix.
4
Example 1.6:
1. The following augmented matrices are in reduced row-echelon form.

2. The following augmented matrices are in row-echelon form but not in reduced row-echelon
form.

3. The following augmented matrices are not in row-echelon form. Hence, the matrices are also
not in reduced row-echelon form.

Remark 1.4:
When the augmented matrix of a system of linear equations is in row-echelon form or reduced
row-echelon form, we can get the solutions of the system easily.

Example 1.8
⎡ 1 0 0 2 ⎤
⎢ ⎥
1. The array ⎢ 0 −1 0 3 ⎥ is the augmented matrix of a system of linear equations in 3
⎢ 0 0 2 4 ⎥
⎣ ⎦
variables x1, x2, x3:
x1 =2
− x2 =3
2x3 = 4.
The system has only one solution: x1 = 2, x2 = −3 , x3 = 2.

⎡ 0 2 2 1 −2 2 ⎤
⎢ ⎥
2. The array ⎢ 0 0 1 1 1 3 ⎥ is the augmented matrix of a system of linear equations
⎢ 0 0 0 0 2 4 ⎥
⎣ ⎦
in 5 variables x1, x2, x3, x4, x5:
2x2 + 2x3 + x4 − 2x5 = 2
x3 + x4 + x5 = 3
2x5 = 4.
Note that x1 does not appear in the equations and this means x1 is arbitrary.
By the third equation x5 = 2. Substitute x5 = 2 into the second equation, we have x3 = 1 − x4.
Substitute x3 = 1 − x4 and x5 = 2 into the first equation, we have x2 = 2 + 12 x4 . (This method
is called the back-substitution.)
5
 Let x1 = s and x4 = t where s, t are arbitrary parameters.
x1 = s
x2 = 2 + 12 t
Then, the general solution is x = 1− t
3 , s, t ∈ R,
x4 = t
x5 = 2
and this system has infinitely many solutions.

⎡ 1 −1 0 3 -2 ⎤
⎢ ⎥
3. The array ⎢ 0 0 1 2 5 ⎥ is the augmented matrix of a system of linear equations in
⎢ 0 0 0 0 0 ⎥
⎣ ⎦
4 variables x1, x2, x3, x4: x1 − x2 + 3x4 = −2
x3 + 2x4 = 5
0 = 0.
Let x2 = s and x4 = t where s, t are arbitrary parameters.
x1 = −2 + s − 3t
x2 = s
Then, the general solution is , s, t ∈ R
x3 = 5 − 2t
x4 = t
and this system has infinitely many solutions.

⎡ 3 1 4 ⎤
⎢ ⎥
4. The array ⎢ 0 2 1 ⎥ is the augmented matrix of a system of linear equations in 2 variables
⎢ 0 0 1 ⎥
⎣ ⎦
x1, x2: 3x1 + x2 = 4
2x2 = 1
0 = 1.
This system is inconsistent. That is, the system has no solution.

1.3 Gaussian Elimination

Algorithm 1.1 ( Gaussian Elimination)

The following procedures reduce a matrix to a row echelon form by some elementary row
operations.

Step 1: Locate the leftmost nonzero column (a column with at least one nonzero entries.)

Step 2: Interchange the top row with another row, if necessary, to bring a nonzero entry to the top
of the column found in Step 1.

Step 3: For each row below the top row, apply some elementary row operations to the rows
below so that the entries below the leading entry of the top row become zeros.

Step 4: Cover the top row in the matrix and apply Step 1 to Step 3 to the submatrix that remains.
Continue in this way until the entire matrix is in row-echelon form.
6
Remark 1.5:
We may write a system of linear equations in its augmented matrix [A | B], then apply Gauss
elimination to reduce the augmented matrix to a row-echelon form, which we shall then readily
obtain the values of the unknowns by “back substitution”.

Algorithm 1.2 ( Gaussian-Jordan Elimination)

Given a matrix, use Gauss Elimination (Algorithm 1.1) to reduce it to a row-echelon form. Then
follow the following procedures reduce it to reduced row-echelon form:

Step 5: Multiply a suitable constant to each row so that all leading entries become 1.

Step 6: Beginning with the last nonzero row and working upward, apply some elementary row
operation Ri → Ri + β R j , β ∈ R to the row above to introduce zero above the leading
entries.

Example 1.9:
⎡ 0 0 2 4 2 8 ⎤
⎢ ⎥
Use Gaussian Elimination to reduce the matrix ⎢ 1 2 4 5 3 −9 ⎥ to a row-echelon
⎢ −2 −4 −5 −4 3 6 ⎥
⎣ ⎦
form.

Solution:
Step 1: The first column is the leftmost nonzero column.
Step 2: Interchange the first and second rows.
⎡ 0 0 2 ⎡ 1 2 4 5 3 −9 ⎤
4 2 8 ⎤ R1↔R2
⎢
⎢ 1 2 4
⎥
5 3 −9 ⎥ ⎯ ⎯⎯→ ⎢⎢ 0 0 2 4 2 8 ⎥
⎥
⎢ −2 −4 −5 −4 3 6 ⎥ ⎢ −2 −4 −5 −4 3 6 ⎥
⎣ ⎦ ⎣ ⎦
Step 3: Use an elementary row operations to the third row so that the entry (3, 1) is zero
R3 →R3 +2 R1 ⎡ 1 2 4 5 3 −9 ⎤
⎢ ⎥
⎯ ⎯⎯⎯→ ⎢ 0 0 2 4 2 8 ⎥
⎢ 0 0 3 6 9 −12 ⎥
⎣ ⎦
Step 4: Cover the first row and apply Step 1 to Step 3 to the submatrix that remains.
Step 1.1: The third column is the leftmost nonzero column.
Step 2.1: No action is need.
Step 3.1: Use an elementary row operations to the third row so that the entry (3, 3) is zero.
R3 →R3 − 23 R2 ⎡ 1 2 4 5 3 −9 ⎤
⎢ ⎥
⎯ ⎯⎯⎯→ ⎢ 0 0 2 4 2 8 ⎥
⎢ 0 0 0 0 6 −24 ⎥
⎣ ⎦
Step 4.1: The matrix is already in row-echelon form.

7
Example 1.10:
Use Gaussian-Jordan Elimination to reduce the matrix in Example 1.9 to a reduced row-echelon
form.
Solution:
First, we follow the steps in Example 4.9 to obtain a row-echelon form as shown in Step 3.1.
Step 5: Multiply a suitable constant to each row so that all leading entries (the entries (1, 1), (2, 3),
and (3, 5)) become 1.
R2 → 12 R21 ⎡ 1 2 4 5 3 −9 ⎤
⎢ ⎥
⎯ ⎯⎯→ ⎢ 0 0 1 2 1 4 ⎥
⎢ 0 0 0 0 1 −4 ⎥
⎣ ⎦
Step 6: Apply some elementary row operations to the first and second rows to introduce zero to
the entries above the leading entry of the third row. That is, introduce zero to the entries
(1, 5) and (2, 5).
R1→R1 −3R31
⎡ 1 2 4 5 0 3 ⎤
⎢ ⎥
⎯ ⎯⎯⎯→ ⎢ 0 0 1 2 0 8 ⎥
⎢ 0 0 0 0 1 −4 ⎥
⎣ ⎦
Repeat Step 6, apply an elementary row operation to the first row to introduce zero to the
entry above the leading entry of the second row. That is, introduce zero to the entries (1, 3).
R1→R1 −4 R21
⎡ 1 2 0 −3 0 −29 ⎤
⎢ ⎥
⎯ ⎯⎯⎯→ ⎢ 0 0 1 2 0 8 ⎥.
⎢ 0 0 0 0 1 −4 ⎥
⎣ ⎦
The matrix is now in reduced row-echelon form.

Example 1.11:
Solve the system of linear equations
2x3 + 4x4 + 2x5 = 8
x1 + 2x2 + 4x3 + 5x4 + 3x5 = −9
−2x1 − 4x2 − 5x3 − 4x4 + 3x5 = 6.
Solution:
⎡ 0 0 2 4 2 8 ⎤
⎢ ⎥
The augmented matrix of a system of linear equations is ⎢ 1 2 4 5 3 −9 ⎥.
⎢ −2 −4 −5 −4 3 6 ⎥
⎣ ⎦

Method 1: Use Gaussian Elimination to reduce the augmented matrix to row-echelon form.
From Example 1.9, a row-echelon form of the above augmented matrix is
⎡ 1 2 4 5 3 −9 ⎤
⎢ ⎥
⎢ 0 0 2 4 2 8 ⎥
⎢ 0 0 0 0 6 −24 ⎥
⎣ ⎦
which corresponds to the system: x1 + 2x2 + 4x3 + 5x4 + 3x5 = −9
2x3 + 4x4 + 2x5 = 8
6x5 = −24.
By the method of back-substitution, we have the third equation, x5 = −4.

8
Substitute x5 = −4 into the second equation, we have 2x3 + 4x4 + 2(−4) = 8 ⇔ x3 = 8 − 2x4.
Substitute x5 = −4 and x3 = 8 − 2x4 into the first equation, we have
x1 + 2x2 + 4(8 − 2x4) + 5x4 + 3(−4) = −9 ⇔ x1 = −29 − 2x2 + 3x4 .
Let x2 = s and x3 = t where s, t are arbitrary parameters.
x1 = −29 − 2s + 3t
x2 = s
Then, the general solution is x = 8 − 2t
3 , s, t ∈ R.
x4 = t
x5 = −4

Method 2: Use Gaussian-Jordan Elimination to reduce the augmented matrix to reduced row-
echelon form. From Example 1.10, the reduced row-echelon form of the augmented matrix is
⎡ 1 2 0 −3 0 −29 ⎤
⎢ ⎥
⎢ 0 0 1 2 0 8 ⎥.
⎢ 0 0 0 0 1 −4 ⎥
⎣ ⎦
which corresponds to the system:
x1 + 2x2 + − 3x4 = −29
x3 + 2x4 =8
x5 = −4.
By the method of back-substitution, we have x5 = −4, x3 = 8 − 2x4 and x1 = −29 − 2x2 + 3x4 .
Let x2 = s and x3 = t where s, t are arbitrary parameters.
x1 = −29 − 2s + 3t
x2 = s
Then, the general solution is x = 8 − 2t
3 , s, t ∈ R.
x4 = t
x5 = −4

Example 1.12:
Determine whether the following system of linear equations is consistent or inconsistent:
3x1 + 2x2 + x3 = 3
2x1 + x2 + x3 = 0
6x1 + 2x2 + 4x3 = 6.

Solution:
The augmented matrix of the system of linear equations is
⎡ 3 2 1 3 ⎤ R2 →R2 − 23 R1
⎡3 2 1 3⎤ ⎡ 3 2 1 3 ⎤
R3 →R3 −2 R1 R3 →R3 −6 R2
⎢ ⎥ ⎢0 − 1 − 1 − 2 ⎥ ⎢ ⎥
⎢ 2 1 1 0 ⎥ ⎯ ⎯⎯⎯→ ⎢ 3 3 ⎥
⎯ ⎯⎯⎯→ ⎢ 0 − 13 − 13 −2 ⎥
⎢ 6 2 4 6 ⎥ ⎢⎣0 − 2 2 0 ⎥⎦ ⎢ ⎥
⎣ ⎦ ⎣ 0 0 0 12 ⎦

From the third row, we have 0(x3) = 12.

However, 0 ≠ 12. Therefore, the system has no solution.
Hence, the system is inconsistent.

9
Alternative method: (To avoid fraction)
⎡ 3 2 1 3⎤ ⎡ 1 1 0 3⎤ R2 →R2 −2 R1 ⎡1 1 0 3 ⎤
⎢2 1 1 0⎥ ⎯⎯ R1 →R1 − R2 R3 →R3 −6 R1
⎢ ⎥
⎢ ⎥
⎯⎯→ ⎢2 1 1 0⎥ ⎯⎯⎯⎯→ ⎢⎢0 − 1 1 6 ⎥⎥
⎣⎢6 2 4 6⎦⎥ ⎣⎢6 2 4 6⎦⎥ ⎣⎢0 − 4 4 12 ⎦⎥
⎡ 1 1 0 3 ⎤
R3 → R3 −4R2 ⎢ ⎥
⎯⎯ ⎯ ⎯ ⎯→ ⎢ 0 −1 1 6 ⎥.
⎢ 0 0 0 −12 ⎥
⎣ ⎦
From the third row, we have 0(x3) = −12.
However, 0 ≠ −12. Therefore, the system has no solution.
Hence, the system is inconsistent.

Remark 1.6:
1. A system of linear equations is inconsistent (that is it has no solution), if a row-echelon form
of its augmented matrix has a row with nonzero last entry but zero elsewhere:

Non-zero

where ⊗ represents the leading entries and is a nonzero number.

⎡ 1 2 4 2 ⎤
⎢ ⎥
For example, ⎢ 0 −1 0 3 ⎥ is a row-echelon form of the augmented matrix of an
⎢ 0 0 0 4 ⎥
⎣ ⎦
inconsistent system.

2. A system of linear equations has exactly one (unique) solution, if a row-echelon form of its
augmented matrix has a leading entry in every column, except the last column:

where ⊗ represents the leading entries.

In other words, a consistent linear system has exactly one solution if the number of variables
in the linear system is equal to the number of nonzero rows in the row-echelon form of its
augmented matrix.
⎡ 1 1 2 3 4 ⎤
⎡ 1 2 4 2 ⎤ ⎢ ⎥
⎢ ⎥ ⎢ 0 2 0 1 −1 ⎥
For example, ⎢ 0 −1 0 3 ⎥ and 0 0 4 −1 2 ⎥ are augmented matrix of some
⎢
⎢ 0 0 2 4 ⎥ ⎢ ⎥
⎢ 0 0 0 −1 2 ⎥
⎣ ⎦
⎢⎣ 0 0 0 0 0 ⎥⎦
consistent systems with exactly one solution.
10
3. A system of linear equations has infinitely many solution, if a row-echelon form of its
augmented matrix has a column which is not the last column and does not consist of leading
entry:

a column (not the last column) with no leading entry

where ⊗ represents the leading entries.
In other words, a consistent linear system has infinitely many solution if the number of
variables in the linear system is greater than the number of nonzero rows in the row-echelon
form of its augmented matrix.

⎡ 1 −1 0 3 0 ⎤ ⎡ 1 2 0 3 0 ⎤
⎢ ⎥ ⎢ ⎥
For example, ⎢ 0 0 1 2 5 ⎥ and ⎢ 0 −1 0 0 5 ⎥ are augmented matrix of some
⎢ 0 0 0 1 3 ⎥ ⎢ 0 0 0 1 3 ⎥
⎣ ⎦ ⎣ ⎦
consistent systems with infinitely many solution.

Example 1.13:
What is the condition that must be satisfied by a, b and c so that the following system of linear
equations is consistent?
x + 2y − 3z = a
2x + 6y − 11z = b
x − 2y + 7z = c

Solution:
The augmented matrix of the system of linear equations is
R2 →R2 −2 R1
⎡ 1 2 −3 a ⎤ R3 →R3 −R1
⎡ 1 2 −3 a ⎤ R3 →R3 +2 R2
⎢ ⎥ ⎢ ⎥
⎢ 2 6 −11 b ⎥ ⎯ ⎯⎯⎯→ ⎢ 0 2 −5 b − 2a ⎥ ⎯ ⎯⎯⎯→
⎢ 1 −2 7 c ⎥ ⎢ 0 −4 10 c − a ⎥
⎣ ⎦ ⎣ ⎦
⎡ 1 2 −3 a ⎤
⎢ ⎥
⎢ 0 2 −5 b − 2a ⎥ .
⎢ 0 0 0 2b + c − 5a ⎥
⎣ ⎦
∴ The system is consistent if and only if 2b + c − 5a = 0.

Example 1.14:
Determine the values of k so that the following system of linear equations

x + 2y + z = 1
2x + ky + 2z = 2
4x + 8y + k2z = 2k

has (a) no solution (b) a unique solution (c) infinitely many solutions.
11
Solution:
The augmented matrix of the system of linear equations is
R2 →R2 −2 R1
⎡ 1 2 1 1 ⎤ R3 →R3 −4 R1
⎡ 1 2 1 1 ⎤
⎢ ⎥ ⎢ ⎥
⎢ 2 k 2 2 ⎥ ⎯ ⎯⎯⎯→ ⎢ 0 k −4 0 0 ⎥.
⎢ 4 8 k2 2k ⎥ ⎢ 0 0 2
k − 4 2k − 4 ⎥
⎣ ⎦ ⎣ ⎦

(a) The system has no solution if k2 − 4 = 0 and 2k − 4 ≠ 0. That is, k = −2.

(b) The system has a unique solution if k − 4 ≠ 0 and k2 − 4 ≠ 0. That is, k ≠ ± 2 and k ≠ 4.
infinitely many solutions
(c) The system has a unique solution if k − 4 = 0 or both k2 − 4 = 0 and 2k − 4 = 0.
That is, k = 2 or k = 4.

1. 4 Homogeneous system (B = 0)
Definition 1.7. A system of m linear equations in n unknowns x1, x2,…, xn is is said to be
homogeneous if it has the form
a11x1 + a12x2 + … + a1nxn = 0
a21x1 + a22x2 + … + a2nxn = 0
!
am1x1 + am2x2 + … + amnxn = 0
where aij ,1 ≤ i ≤ m, 1 ≤ j ≤ n are real constants.
A homogeneous system always has a solution, x1 = 0, x2 = 0, x3 = 0, …, xn = 0. This solution is
called the trivial solution. Any other solution, if it exists, is called nontrivial solution.
Remark 1.8:
1. A homogeneous system of linear equations is consistent, it is has either only the trivial solution
or infinitely many solutions in addition to the trivial solution.

2. A homogeneous system of linear equations with more unknowns than equations has infinitely
many solutions.

Example 1.15: Determine the value(s) of k such that x − 5y + 3z = 0

5x − y − kz = 0
x − 2y + kz = 0
has nontrivial solution.
Solution:
The augmented matrix of the system of linear equations is
R2 →R2 −5R1
⎡ 1 −5 3 0 ⎤ R3 →R3 +R1
⎡ 1 −5 3 0 ⎤ R3 →8R32
⎢ ⎥ ⎢ ⎥
⎢ 5 −1 −k 0 ⎥ ⎯ ⎯⎯⎯→ ⎢ 0 24 −k −15 0 ⎥ ⎯ ⎯⎯→
⎢ 1 −2 k 0 ⎥ ⎢ 0 3 k −3 0 ⎥
⎣ ⎦ ⎣ ⎦
⎡ 1 −5 3 0 ⎤ R3 →R3 −R22
⎡ 1 −5 3 0 ⎤
⎢ ⎥ ⎢ ⎥
⎢ 0 24 −k −15 0 ⎥ ⎯ ⎯⎯⎯ → ⎢ 0 24 −k −15 0 ⎥
⎢ 0 24 8k − 24 0 ⎥ ⎢ 0 0 9k − 9 0 ⎥
⎣ ⎦ ⎣ ⎦
∴ The system has nontrivial solution if 9k − 9 = 0, that is k = 1.

(Remark: If k ≠ 1, then the system has only trivial solution x = y = z = 0.)

12