Advanced Engineering Mathematics

FOURIER SERIES, INTEGRALS AND TRANSFORMS

Fourier series are infinite series designed to represent general periodic functions in terms of cosines and
sines. They constitute a very important tool, in particular in solving problems that involve ODEs and PDEs.
The theory of Fourier series is complicated, but we shall see that the application of these series is rather
simple. Fourier series are in a certain sense more universal than Taylor series in calculus because many
discontinuous periodic functions of practical interest can be developed in Fourier series but do not have
Taylor series representations.

In the latter part, we extend the ideas and techniques of Fourier series to nonperiodic functions which give
rise to Fourier integrals and Fourier transforms. These have applications to solving PDEs.

3.1 Fourier Series

applications. A function ft is called a periodic function if ft is defined for all real t (except perhaps for
Fourier series are the basic tool for representing periodic functions, which play an important role in

some points) and if there is some positive number p, called a period of ft such that

ft + p = ft (3.1)

The graph of such a function is obtained by periodic repetition of its graph in any interval of length p.

If ft has period p, it also has the period 2p. Thus for any integer n = 1, 2, 3 …

ft + np = ft (3.2)

Furthermore, if ft and gt have period p, then aft + bgt with any constants a and b also has
period p.

We will represent the periodic function ft of period 2π in terms of simple functions whose period is also
2π. The series to be obtained will be a trigonometric series of the form

a + a cos t + b sin t + a cos 2t + b sin 2t + ⋯

Thus, the Fourier series expansion of a periodic function ft with p = 2π is



ft = a + a cos nt + b sin nt (3.3)



provided the right term converges to ft. The constants a , a , b , a , b , etc., are called the coefficients
of the series or the Fourier coefficients given by the following Euler formulas

Fourier Series, Integrals and Transforms Page 1

Advanced Engineering Mathematics

1
a =  ft dt
2π
(3.4a)
!

1
a =  ft cos nt dt
π
(3.4b)
!

1
b =  ft sin nt dt
π
(3.4c)
!

where n = 1, 2, 3, ⋯ and the limits of the integral defined for the whole period of ft.

Example 3.1

−k − π < t < 0
Find the Fourier coefficients of the periodic function
ft = "
k 0 < t < π

and determine its Fourier series expansion.

Answer:

a = 0
Fourier coefficients:

a = 0
2k
b = '1 − −1 (
nπ
4k
b = b = 0
π
4k
b* = b+ = 0
3π
4k
b, = b. = 0
5π
4k
b/ = b1 = 0
7π
4k
b2 = b = 0
9π
⋮ ⋮

4k 1 1 1
Fourier series expansion:
f  t = 5sin t + sin 3t + sin 5t + sin 7t + ⋯ 6
π 3 5 7

Fourier Series, Integrals and Transforms Page 2

Advanced Engineering Mathematics

Figure 3.1. The given periodic function f(t). Periodic rectangular wave.

Figure 3.2. Partial sums of the Fourier series expansion for f(t)

Fourier Series, Integrals and Transforms Page 3

Advanced Engineering Mathematics

Find the Fourier series expansion of the given ft which is assumed to have the period 2π. In the
Drill Problems 3.1

expansion, terms involving cos 5t and sin 5t must be included.

1.

2.

3.

4.

5.

Fourier Series, Integrals and Transforms Page 4

Advanced Engineering Mathematics

6.

7.

8.

9. ft = t  −π < t < π

10. ft = t  0 < t < 2π

3.2. Functions of any Period p = 2L

The functions considered so far had period 2π, for the simplicity of formulas. However, periodic functions of

expansion of function ft whose period is p = 2L is

practical interest will generally have other periods. By change of scale, we find that the Fourier series


πn πn
ft = a +  8a cos t + b sin t9
L L
(3.5)


and the Fourier coefficients are found by the formulas

Fourier Series, Integrals and Transforms Page 5

Advanced Engineering Mathematics

1
a =  ft dt
2L
(3.6a)
!

1 πn
a =  ft cos t dt
L L
(3.6b)
!

1 πn
b =  ft sin t dt
L L
(3.6c)
!

Find the Fourier series expansion of the function whose period is p = 4

Example 3.2

−k − 2 < t < 0
ft = "
k 0 < t < 2

4k πt 1 3πt 1 5πt 1 7πt

Answer
ft = 5sin + sin + sin + sin + ⋯6
π 2 3 2 5 2 7 2

Find the Fourier series expansion of the following periodic functions. Terms involving cos 5t and sin 5t
Drill Problems 3.2

must be included.

−1 − 2 < t < 0
1. ft = " ,p= 4
1 0 < t < 2
0 − 2 < t < 0
2. ft = " ,p =4
4 0 < t < 2

3. ft = t  − 1 < t < 1, p = 2

4. ft = − 1 < t < 1, p = 2

:;<


5. ft = sin πt 0 < t < 1, p = 1

6. ft = cos πt −  < t < , p = 1

 

7. ft = |t| − 1 < t < 1, p = 2

Fourier Series, Integrals and Transforms Page 6

Advanced Engineering Mathematics

1 + t − 1 < t < 0
8. ft = " ,p =2
1 − t 0 < t < 1

9. ft = 1 − t  − 1 < t < 1, p = 2

0 − 2 < t < 0
10. ft = " ,p=4
t 0 < t < 2

3.3 Even and Odd Functions.

A function ft is even if f−t = ft and is symmetric with respect to the vertical axis.

Figure 3.3. An even function.

A function ft is odd if f−t = −ft and is symmetric with respect to the origin.

Figure 3.4. An odd function

Fourier Series, Integrals and Transforms Page 7

Advanced Engineering Mathematics

The Fourier series of an even function of period p = 2L is a Fourier cosine series


πn
ft = a +  a cos t
L
(3.7)


with coefficients (integration from 0 to L only!)

>
1
a =  ft dt
L
(3.8a)


>
2 πn
a =  ft cos t dt
L L
(3.8b)


The Fourier series of an odd function of period p = 2L is a Fourier sine series


πn
ft =  b sin t
L
(3.9)


with coefficients
>
2 πn
b =  ft sin t dt
L L
(3.10)


The function in Examples 3.1 and 3.2 is an odd function. Therefore, it has a Fourier sine series expansion
given by 3.9 and coefficients given by 3.10.

Fourier Series, Integrals and Transforms Page 8

Advanced Engineering Mathematics

Example 3.3

0 − 2 < t < −1
Find the Fourier series expansion of the function

ft = ? k − 1 < t < 1

0 1 < t < 2

whose period is p = 2L = 4

k 2k πt 1 3πt 1 5πt
Answer
ft = + 5cos − cos + cos − ⋯+ ⋯6
2 π 2 3 2 5 2

Figure 3.5. The function f(t) in Example 3.3

Drill Problems 3.3

involving cos 5t and sin 5t must be included.

Determine whether the following functions are even or odd then find its Fourier series expansion. Terms

1. ft = π − |t| − π < t < π

2. ft = 2t|t| − 1 < t < 1

t −  < t <
: :

3. ft = ? 
π−t <t<
: *:
 

4. ft = "πe ; −π < t < 0

A;

πe 0 < t < π
2 −2 < t < 0
5. ft = "
0 0 < t < 2

3.4 Fourier Integral

Fourier series are powerful tools for problems involving functions that are periodic or are of interest on a
finite interval only. Many problems involve functions that are nonperiodic and are of interest on the whole x-
axis, thus, we extend the method of Fourier series to such functions. This idea will lead to Fourier integrals.

Fourier Series, Integrals and Transforms Page 9

Advanced Engineering Mathematics

We will start from a function whose period is p = 2L, derive its Fourier series expansion, and then let
L → ∞. For a rectangular pulse, Figure 3.6 illustrates the idea

Figure 3.6. Waveform and amplitude spectrum of a rectangular pulse

For a periodic function f> t whose period is p = 2L, its Fourier series expansion is given by


f> t = a + a cos w t + b sin w t (3.11)



where w = . If we let L → ∞, then we will haveA, for nonperiodic ft,

:
>


ft =  'Aw cos wt + Bw sin wt(dw (3.12)


Fourier Series, Integrals and Transforms Page 10

Advanced Engineering Mathematics

with

1 
Aw =  fv cos wv dv
π A
(3.13a)

1 
B w =  fv sin wv dv
π A
(3.13b)

Equation 3.12 is the Fourier integral representation of ft . It can be seen that this representation is
continuous, unlike the Fourier series for periodic function which is defined for integer values of n, and thus
the quantity w = becomes continuous also.
:
>

becomes simpler. If ft is an even function, then Bw = 0 in 3.13b and

Fourier Cosine Integral and Fourier Sine Integral. For an even or odd function, the Fourier integral

1 
Aw =  fv cos wv dv
π A
(3.13a)

and the Fourier integral then reduces to the Fourier cosine integral. Similarly if ft is an odd function,
Aw = 0 in 3.13a and

1 
Bw =  fv sin wv dv
π A
(3.13b)

The Fourier integral becomes the Fourier sine integral.

3.5 Fourier Transform

The Fourier cosine and sine transforms are real. We now consider a third one, called the Fourier transform,
which is complex. From the real form of the Fourier integral

ft =  'Aw cos wt + Bw sin wt(dw (3.12)


with

1 
Aw =  fv cos wv dv
π A
(3.13a)

Fourier Series, Integrals and Transforms Page 11

Advanced Engineering Mathematics

1 
Bw =  fv sin wv dv
π A
(3.13b)

Substituting these into the integral, we have

1  
ft =   fv'cos wv cos wt + sin wv sin wt(dv dw
π  A
(3.14)

The terms inside the bracket is the addition formula for cosine, or coswx − wv. Thus,

1  
ft =  I fv coswx − wv dv J dw
π 
(3.15)
A

We shall call the term inside the bracket Fw since coswx − wv is an even function of w, and we
integrate with respect to v. We extend the limit of the integral of Fw from w = 0 to ∞ to w = −∞ to
+∞; thus a factor of  is multiplied to the integral to account for this extension. Thus,


1  
ft =  I fv coswx − wv dv J dw
2π A A
(3.16)

We claim that the integral of 3.16 with sine instead of a cosine is zero,

1  
 I fv sinwx − wv dv J dw = 0
2π A A
(3.17)

This is true since sinwx − wv is an odd function of w, which makes the integral inside the bracket,
calling it Gw, equal to zero.

Taking the integrand of 3.16 plus j times the integrand of 3.17 and use the Euler’s formula

eN; = cos t + j sin t (3.18)

Thus, the complex Fourier integral can now be written as

1  
ft =   fveNOPAQ dv dw
2π A A
(3.19)

Fourier Series, Integrals and Transforms Page 12

Advanced Engineering Mathematics

Writing 3.19 as a product of exponentials

1 
1 
ft =  I  fveANOQ dvJ eNO; dw
√2π A √2π A
(3.20)

The expression in brackets is a function of w - for application purposes, we let w = ω – is denoted by

Fω and is called the Fourier transform of ft; writing v = t. Thus,

1 
F ω  =  fteANO; dt
√2π
(3.21)
A

With this, 3.20 becomes

1 
ft =  Fw eNO; dw
√2π
(3.22)
A

which is the inverse Fourier transform of Fw. We will use the notation

ℱ 'ft( = Fω (3.23a)

to indicate the Fourier transform of ft and

ℱ A 'Fω( = ft (3.23b)

as the inverse Fourier transform of Fω.

Find the Fourier transform of ft = 1 if |x| < 1 and ft = 0 otherwise.
Example 3.4

π sin ω
Answer:
F ω  = U
2 ω

Find the Fourier transform of ft = eAV; for t > 0 and ft = 0 for t < 0.
Example 3.5

1
Answer:
F ω  =
√2πa + jω

Fourier Series, Integrals and Transforms Page 13

Advanced Engineering Mathematics

Drill Problems 3.4

Find the Fourier transform of the following functions using the Fourier transform integral.

||
1. ft = X 1 − V t ≤ a
|;|

0 otherwise

2. ft = eAV|;|

3. ft = "te − 1 < t < 0

A;

0 otherwise
|t| − 1 ≤ t ≤ 1
4. ft = "
0 otherwise

1 0 ≤ t ≤ a/2
5. ft = ?−1 − a/2 ≤ t ≤ 0
0 otherwise

3.6 Properties of Fourier Transform

Linearity. Let ft and gt be two functions with Fourier transforms Fω and Gω respectively. Then,
for constants a and b,

ℱ 'aft + bgt( = aFω + bGω (3.24)

Shift in the frequency domain. Let ft be a function with Fourier transform Fω. Then

ℱ]eNV; ft^ = Fω − a (3.25)

If the function ft has the Fourier transform Fω, use the linearity and shifting in the frequency domain
Example 3.6

properties of the Fourier transform to find ℱ 'ft cos at(.

Fω − a Fω + a
Answer:
F ω  = +
2 2

Time scaling. Let ft be a function whose Fourier transform is Fω. For a constant c,
ω
F8c9
ℱ 'fct( =
c
(3.26)

Fourier Series, Integrals and Transforms Page 14

Advanced Engineering Mathematics

A special case of the scaling is called time-reversal, that is c = −1. Equation 3.26 becomes then,

ℱ'f−t( = −F−ω (3.27)

1 − 1 ≤ t ≤ 1
Example 3.7
If ft = " , plot f2t and f 89and find their Fourier transform. Use the results of
;
0 otherwise
Example 3.4.

ω
Answer:
2 sin 2
ℱ 'f2t( = _
π ω

t 2 sin 2ω
ℱ `f 5 6a = _
2 π ω

Differentiation in time domain. Let ft be a continuously differentiable function with Fourier transform
Fω and assume that lim;→± ft = 0. Then,

ℱ 'f e t( = jωFω (3.28)

and in general, if ft can be differentiated m times,

ℱ]f f t^ = jωf Fω (3.29)

This property of Fourier transform can be used in the same manner as the Laplace transform in solving
ordinary differential equations.

Differentiation in frequency domain. If the function ft has the Fourier transform Fω and if t ft is
absolutely integrable, then Fω is differentiable and

ℱ't ft( = −j F e ω (3.30)

and in general

ℱ 'jtf ft( = F f ω (3.31)

Fourier Series, Integrals and Transforms Page 15

Advanced Engineering Mathematics

t − 1 ≤ t ≤ 1
Example 3.8
Find the Fourier transform of ft = "
0 otherwise
, using the results of Example 3.4 and the
differentiation property of the Fourier transform.

Answer:
2 ω cos ω − sin ω
F ω  = _
π ω

Integration in time-domain. If ft is a continuous and absolutely integrable function with Fourier
transform Fω, and lim;→ gA fτ dτ = 0. Then
;

;
F ω 
ℱ I fτ dτJ =
jω
(3.32)
A

for ω ≠ 0.

The following table summarizes the Fourier transform of various functions.

ft ℱ 'ft( = Fω

2 sin bω
" 1 − b < t < b _
0 otherwise π ω
3.1.1

eANjk − eANlk
"1 b < t < c
0 otherwise jω√2π
3.1.2

1 π eAV|k|
a > 0 U
x + a 2 a
3.1.3
t 0 < t < b −1 + 2eNjk − eANjk
?2t − b b < t < 2b
0 otherwise ω √2π
3.1.4

1
" e t > 0
AV;
a > 0
0 otherwise √2π a + jω
3.1.5

e VANkl
− eVANkj
" e b < t < c
V;

0 otherwise √2π a − jω

3.1.6

2 sin bω − a
me − b < t < b
NV;
_
0 otherwise π ω − a
3.1.7

j eNjVAk − eNlVAk
m e b < t < c
NV;

0 otherwise √2π a−ω

3.1.8
1 Akn /+V
eAV;
n
a > 0 e
√2π
3.1.9

Fourier Series, Integrals and Transforms Page 16

Advanced Engineering Mathematics

ft ℱ 'ft( = Fω

π
sin at U |ω| < a
a > 0 o 2
t
0 |ω| > a
3.1.10

Table 3.1. Table of Fourier Transforms

The following table summarizes the Fourier transform of distributed functions, i.e. functions that are not
square-integrable but have Fourier transforms.

ft ℱ 'ft( = Fω Remarks

1 √2πδω
The transform of a constant
3.2.1 function is the Dirac delta
function.
1
δt
The dual of rule 3.2.1. The
√2π
3.2.2 transform of the Dirac delta
function is a constant.
eNV; √2π δω − a
Applying frequency shift to
3.2.3
δω − a + δω + a
3.2.1.
cos at √2π
2
Using rules 3.2.1, 3.2.3 and
3.2.4
δω + a + δω − a
the Euler’s formula for cosine
sin at j√2π
2
Using rules 3.2.1, 3.2.3 and
3.2.5
the Euler’s formula for sine
Table 3.2. Table of Fourier Transform for distributed functions

Drill Problems 3.5

Prove rules 3.1.1 through 3.1.10 using the integral for the Fourier transform and its properties.

Fourier Series, Integrals and Transforms Page 17