ADVANCED ENGINEERING MATHEMATICS

010.141

FOURIER SERIES, INTEGRALS,

AND TRANSFORMS

MODULE 3
 Fourier Series, Integrals, and Transform

Ø Fourier series: Infinite series designed to represent general periodic

functions in terms of simple ones (e.g., sines and cosines).

Ø Fourier series is more general than Taylor series because many

discontinuous periodic functions of practical interest can be developed
in Fourier series.

Ø Fourier integrals and Fourier Transforms extend the ideas and

techniques of Fourier series to non-periodic functions and have basic
applications to PDEs.

B.D. Youn
2010 Engineering Mathematics II Module 2 2
 PERIODIC FUNCTIONS

Ø A periodic function is a function such that f(x + p) = f(x)

where p is a period.
Ø The smallest period is called a fundamental period.

B.D. Youn
2010 Engineering Mathematics II Module 2 3
 FOURIER SERIES

Assume that f(x) is periodic with period 2p and is integrable

over a period.

f(x) can be represented by a trigonometric series:

¥
f ( x) = a 0 + å (a n cos nx + b n sin nx)
n=1

i.e., the series converges and has sum f(x)

B.D. Youn
2010 Engineering Mathematics II Module 2 4
 TRIGONOMETRIC SERIES WITH A PERIOD OF 2p
Euler Formulas

Fourier coefficients
p
1
a0 =
2p òp f ( x ) dx
-
p
1
an =
p òp f ( x ) × cos nx dx
-
p
1
bn =
p òp f ( x ) × sin nx dx
-

a0, an, bn are the Fourier coefficients.

The corresponding series is the Fourier series
B.D. Youn
2010 Engineering Mathematics II Module 2 5
 FOURIER SERIES (cont)
Example: Rectangular wave

ì - k, - p < x < 0
f ( x) = í
î k, 0 < x < p

Fourier Coefficients p
1
1
0
1
p bn =
p òp f ( x ) sin nx dx
-
a0 = òp -k dx+ ò k dx = 0
2p -
2p 0 1
0
1
p

1
p
1
=
p òp ( -k ) sin nx dx + p ò k sin nx dx
- 0
an = ò f ( x ) cos nx dx= [0] =0
p -p
p k æ é cos nx ù
0
é cos nx ù
p
ö
= çê ú + ê- ÷÷
ç
p è ë n û -p ë n úû 0 ø
k 4k
= (1 + 1 + 1 + 1) = for odd n
pn np
B.D. Youn
2010 Engineering Mathematics II Module 2 6
 EXAMPLE 1: RECTANGULAR WAVE

Fourier series:

¥ ¥
f ( x ) =a 0 + å a n cos nx + å bn cos nx
n=1 n=1
¥
= å b2n+1 sin ( 2n + 1) x
n=0
¥
4k
=å sin ( 2n + 1) x
n=0 ( 2n + 1) p
4k æ 1 1 ö
= sinx+ sin 3 x + sin 5 x + L
p çè 2 ×1+1 5
÷
ø

B.D. Youn
2010 Engineering Mathematics II Module 2 7
 ORTHOGONALITY OF SINE, COSINE FUNCTIONS

Two functions fm and fn of the same form are orthogonal if ò fm fn dx = 0 for

all m ¹ n and ò fm fn dx = a for all m = n.

ò cos mx cos nx dx = 0 for all m ¹ ± n

- p
p
2

-p
ò ( cos nx ) dx ¹ 0 for all m = n

ò sin mx sin nx dx = 0 for all m ¹ n

- p
p
2

-p
ò ( sin nx ) dx ¹ 0 for all m = n

ò cos mx sin nx dx = 0 for all m = n

- p

B.D. Youn
2010 Engineering Mathematics II Module 2 8
 CONVERGENCE AND SUM OF FOURIER SERIES

B.D. Youn
2010 Engineering Mathematics II Module 2 9
 CONVERGENCE AND SUM OF FOURIER SERIES

Proof: For continuous f(x) with continuous first and second order
derivatives.

p
1
an = ò f ( x ) cos nx dx
p - p

p ¢ p
1 æ f ( x ) sin nx ö 1 sin nx
= ò ç ÷ dx - ò f¢ dx
p -p è n ø p -p
n
p p
1 f ( x ) sin nx 1
= - ò f 'sin nx dx
p n p n -p
1442443 -p
0
p
1
=- ò f 'sin nx dx
p n -p

B.D. Youn
2010 Engineering Mathematics II Module 2 10
 CONVERGENCE AND SUM OF FOURIER SERIES (cont)

Repeating the process:

p
1
an = -
pn 2 ò f " (x )cos nx dx
-p

Since f" is continuous on [– π, π ]

f " (x ) < M
p p
1 1
an = 2 ò f " (x ) cos nx dx < 2 ò M dx
pn -p pn -p

2pM 2M
an < =
pn 2 n2
B.D. Youn
2010 Engineering Mathematics II Module 2 11
 CONVERGENCE AND SUM OF FOURIER SERIES (cont)

Similarly for
2M
bn <
n2

Hence
æ 1 1 1 1 ö
f (x ) < a 0 + 2M ç1 + 1 + 2 + 2 + 2 + 2 + L÷
è 2 2 3 3 ø
which converges. (see Page 670)

B.D. Youn
2010 Engineering Mathematics II Module 2 12
 FUNCTIONS OF ANY PERIOD
P = 2L

B.D. Youn
2010 Engineering Mathematics II Module 2 13
 FUNCTIONS OF ANY PERIOD (cont)
P = 2L
Proof: Result obtained easily through change of scale.
px
v=
L
¥ ¥
g(v ) = a 0 + å a n cos nv + å b n sin nv
n =1 n =1

p L L
1 1 npx p 1 npx
an = g (v ) cos nv dv = f (x ) cos × dx = f (x ) cos dx
p -òp p -òL L L L ò
-L
L

Same for a0, bn

p L L
1 1 p 1
a0 = g (v ) dv = f (x ) dx = ò f (x )dx
2p -òp 2p -òL L 2L -L

B.D. Youn
2010 Engineering Mathematics II Module 2 14
 ANY INTERVAL (a, a + P)
P = PERIOD = 2L

¥
æ 2npx 2npx ö
f (x ) = a 0 + å çè na cos + b n s in ÷
n =1
P P ø

a+ P
1
a0 =
P ò f (x )dx
a
a+ P
2 2npx
an = ò f (x ) cos dx
P P
a
a+ P
2 2npx a a+P
bn = ò f (x )sin dx
P P
a

B.D. Youn
2010 Engineering Mathematics II Module 2 15
 EXAMPLE OF APPLICATION
Find the Fourier series of the function
f(x)

ì 1 1< x < 4 P=6

ï
f (x ) = í a =1
ï x
î- 1 4<x<7 1< x < 7 1 4 7

Solution:

7 é 4 7 ù
1 1ê
a0 =
6 ò f (x )dx = 6 ê ò 1dx + ò (-1)dx úú = 0
1 ë1 4 û
é 4 7 ù
1ê npx npx ú 1 é 4np np 7np ù
an = ò cos dx - ò cos dx = 2 sin - sin - sin n¹0
3ê 3 3 ú np êë 3 3 3 úû
ë1 4 û
7 é 4 7 ù
1 npx 1ê npx npx ú 1 é 4np np 7np ù
bn = ò f (x ) sin dx = sin dx - sin dx = - 2cos - cos - cos
3 3 3ê ò 3 ò 3 ú np êë 3 3 3 úû
1 ë1 4 û

B.D. Youn
2010 Engineering Mathematics II Module 2 16
 EXAMPLE OF APPLICATION (cont)

4np np 7np np
sin = cos np sin sin = sin
3 3 3 3
4np np 7np np
cos = cos np cos cos = cos
3 3 3 3

a0 = 0 f(x)

æ 1 - cos np ö np
an = - 2ç ÷ sin
è n p ø 3 x
-3 1 4 7
æ 1 - cos np ö np
bn = 2 ç ÷ cos
è np ø 3

B.D. Youn
2010 Engineering Mathematics II Module 2 17
 EVEN AND ODD FUNCTIONS

Examples: x4, cos x

f(x) is an even function of x, if f(-x) = f(x). For example, f(x) = x sin(x), then

f(- x) = - x sin (- x) = f(x)

and so we can conclude that x sin (x) is an even function.

B.D. Youn
2010 Engineering Mathematics II Module 2 18
 EVEN AND ODD FUNCTIONS

Properties
1. If g(x) is an even function

L L

ò g(x ) dx = 2 ò g(x ) dx
-L 0

2. If h(x) is an odd function

L

ò h(x ) dx = 0
-L

3. The product of an even and odd function is odd

B.D. Youn
2010 Engineering Mathematics II Module 2 19
 EVEN AND ODD FUNCTIONS

B.D. Youn
2010 Engineering Mathematics II Module 2 20
 EVEN AND ODD FUNCTIONS

B.D. Youn
2010 Engineering Mathematics II Module 2 21
 SUM

B.D. Youn
2010 Engineering Mathematics II Module 2 22
 SUM

B.D. Youn
2010 Engineering Mathematics II Module 2 23
 SUM

B.D. Youn
2010 Engineering Mathematics II Module 2 24
 SUM

B.D. Youn
2010 Engineering Mathematics II Module 2 25
 Forced Oscillations

B.D. Youn
2010 Engineering Mathematics II Module 2 26
 Forced Oscillations

B.D. Youn
2010 Engineering Mathematics II Module 2 27
 Forced Oscillations

B.D. Youn
2010 Engineering Mathematics II Module 2 28
 Forced Oscillations

B.D. Youn
2010 Engineering Mathematics II Module 2 29
 Forced Oscillations

B.D. Youn
2010 Engineering Mathematics II Module 2 30
 APPROXIMATION BY
TRIGONOMETRIC POLYNOMIALS

Consider a function f(x), periodic of period 2π. Consider an

approximation of f(x),

N
f ( x) » F ( x ) = A0 + å A cos nx + B
n n sin nx
n=1

The total square error of F

p
2
E= ò (f - F ) dx
-p

is minimum when F's coefficients are the Fourier coefficients.

Read Page 503 to see the derivation procedure of Eq. (6)

B.D. Youn
2010 Engineering Mathematics II Module 2 31
 PARSEVAL'S THEOREM

The square error, call it E*, is

Where an, bn are the Fourier coefficients of f.

¥ p
1
2a02 + å (a
n=1
2
n )
+ bn2 £
p ò f 2 ( x ) dx
-p

B.D. Youn
2010 Engineering Mathematics II Module 2 32
 FOURIER INTEGRALS

Since many problems involve functions that are nonperiodic and are of
interest on the whole x-axis, we ask what can be done to extend the method
of Fourier series to such functions. This idea will lead to “Fourier integrals.”

B.D. Youn
2010 Engineering Mathematics II Module 2 33
 FOURIER INTEGRALS

B.D. Youn
2010 Engineering Mathematics II Module 2 34
 FOURIER INTEGRALS

B.D. Youn
2010 Engineering Mathematics II Module 2 35
 FOURIER COSINE AND SINE INTEGRALS

B.D. Youn
2010 Engineering Mathematics II Module 2 36
 FOURIER COSINE AND SINE INTEGRALS

B.D. Youn
2010 Engineering Mathematics II Module 2 37
 FOURIER COSINE AND SINE INTEGRALS

¥
f (x) = ò éë A (w ) cos w x + B (w ) sin w x ùû dw
0
¥
1
A (w ) = ò f ( v ) cos (w v ) dv
p -¥
¥
1
B (w ) = ò f ( v ) sin (w v ) dv
p -¥

¥ p
1
f ( x) = a 0 + å (a n cos nx + b n sin nx) a0 =
2p òp f ( x ) dx
-
n=1
p
1
an =
p òp f ( x ) × cos nx dx
-
p
1
bn =
p òp f ( x ) × sin nx dx
-

B.D. Youn
2010 Engineering Mathematics II Module 2 38
 EXISTENCE

If f(x) is piecewise continuous in every finite interval and has a right hand
and left hand derivative at every point and if

0 b
lim
a ® -¥
ò f (x ) dx + blim
®¥
ò f (x ) dx
a 0

exists, then f(x) can be represented by the Fourier integral.

Where f(x) is discontinuous, the F.I. equals the average of the left-hand
and right-hand limit of f(x).

B.D. Youn
2010 Engineering Mathematics II Module 2 39
 FOURIER COSINE AND SINE INTEGRALS

For an even or odd function the F.I. becomes much simpler.

If f(x) is even
¥
2
A(ω) = ò f (v ) cos ωv dv
π -¥
¥
f (x ) = ò A(ω)cos (ωx ) dω Fourier cosine integral
0

If f(x) is odd
¥
2
B(ω) = ò f (v ) sin ωv dv
π -¥
¥
f (x ) = ò B(ω)sin (ωx ) dω Fourier sine integral
0

B.D. Youn
2010 Engineering Mathematics II Module 2 40
 EXAMPLE

Consider f (x ) = e - kx x > 0, k > 0

Evaluate the Fourier cosine integral A(w) and sine integral B(w).

For Fourier cosine integral,

Integration by parts gives

Fourier cosine integral is

B.D. Youn
2010 Engineering Mathematics II Module 2 41
 EXAMPLE

For Fourier sine integral,

Integration by parts gives

Fourier sine integral is

Laplace integrals

B.D. Youn
2010 Engineering Mathematics II Module 2 42
 FOURIER SINE AND COSINE TRANSFORMS

For an even function, the Fourier integral is the Fourier cosine integral
¥ ¥
2 2 ˆ
f (x ) = ò A(ω)cos (ωx ) dω A(ω) = ò f (v ) cos ωv dv = f c (w )
0
π -¥
π

Then
¥ ¥
π π 2 2
fˆc (w ) = A(ω) = ò f (v) cos ωv dv = ò f (v) cos ωv dv
2 2 π 0
π 0
¥ ¥
2 ˆ 2
f (x ) =
0
ò π
f c (w ) cos ωx dw =
π 0
ò fˆ (w ) cos ωx dw
c

fˆc (w ) is defined as the Fourier cosine transform of f .

f is the inverse Fourier cosine transform of fˆ . c

B.D. Youn
2010 Engineering Mathematics II Module 2 43
 FOURIER SINE AND COSINE TRANSFORMS (cont)

Similarly for odd function

¥
2
F sine T ® f̂ c (w) = ò f (x ) sin ωx dx
π 0

¥
2
Inverse F sine T ® f (x ) = ò f̂ c (w) sin ωx dw
π 0

B.D. Youn
2010 Engineering Mathematics II Module 2 44
 NOTATION AND PROPERTIES

Fc { f } = fˆc , Fs { f } = fˆs
{} {}
Fc-1 fˆc = f , Fs-1 fˆs = f

(1) Fc {af + bg} = a Fc {f } + b Fc {g}

(2) Fs {af + bg} = a Fs {f } + b Fs {g}
2
(3) Fc {f ¢(x )} = ω Fs {f } - f (0 )
π
(4) Fs {f ¢(x )} = - ω Fc {f }
2
(5) Fc {f ¢¢} = - ω 2 Fc {f } - f ¢(0 )
π
2
(6) Fs {f ¢¢} = - ω 2 Fs {f } + ωf ¢(0 )
π

B.D. Youn
2010 Engineering Mathematics II Module 2 45
 FOURIER TRANSFORM
The F.I. is: ¥
f (x ) = ò [A(ω) cos ωx + B(ω) sin ωx ] dx
0
¥
1
A(w) = ò f (v ) cos ωv dv
π -¥
¥
1
B(w) = ò f (v ) sin ωv dv
π -¥
Replacing
¥ ¥
1
f (x ) =
p ò dw ò [f (v ) cos wv cos wx + f (v ) sin wv sin wx ]dv
0 -¥
¥ ¥
1
=
p ò dw ò f (v ) cos (wv - wx )dv
0 -¥
1444424444
3
Even in w

B.D. Youn
2010 Engineering Mathematics II Module 2 46
 FOURIER TRANSFORM

¥ ¥
1
f (x ) = ò dw ò f (v )cos w (x - v )dv
2p -¥ -¥

¥ ¥
1
f (x ) = ò dw ò f (v )sin w (x - v )dvl
2p -¥ -¥
1444
424444
3
odd in w
Now,
¥
1
f (x ) = F(ω) + i G (w) dω
2p - ¥ò

¥ ¥
1 iω(x - v )
= ò dw ò f (v ) e dv Complex Fourier Integral
2p -¥ -¥

B.D. Youn
2010 Engineering Mathematics II Module 2 47
 FOURIER TRANSFORM (cont)

¥ ¥
1 iωx 1 - iωv
f (x ) = ò e dw ò f ( v ) e dv
2p -¥
2p - ¥
144424443
f̂ (w) º Fourier Transform of f

¥
1 iωx
f (x ) = ò f̂ (w ) e dw
2p -¥

B.D. Youn
2010 Engineering Mathematics II Module 2 48
 NOTATION AND PROPERTIES

F {af + bg} = aF {f } + bF {g}

F {f ¢} = iω F {f }

F {f ¢¢} = - ω 2 F {f }

F {f * g} = 2p F {f } + F {g}

B.D. Youn
2010 Engineering Mathematics II Module 2 49