Lecture 1 : Fourier Series

Contents

• Periodic Functions.

• Piecewise Continuous Functions.

• Types of functions.

• Fourier Series.

• Half Range Expansions (Fourier Sine Expansion – Fourier Cosine Expansion).

 What is the Fourier Series?

It’s a mathematical way to represent a periodic (wave-like) function as the sum of

simple sine and cosine waves. It decomposes any periodic function or periodic signal
into the sum of an infinite set of simple oscillating functions, sines and cosines

• Fourier series represents the base of sampling theorem in communication science.

• It’s also could be used in solving some partial differential equations.
 Some important definitions
1. Periodic function
It is a function repeats itself every constant time (called the period T) so that 𝑓(𝑡 + 𝑇) = 𝑓(𝑡)
 Periodic function
period=2𝐿
FOURIER 𝒂‫ه‬ 𝒂𝒏 𝒃𝒏
COEFFICIEN
T
GENERAL 1 𝐿 1 𝐿 1 𝐿
CASE න 𝑓 𝑥 𝑑𝑥 න 𝑓 𝑥 cos 𝑛𝑥 𝑑𝑥 න 𝑓 𝑥 sin 𝑛𝑥 𝑑𝑥
𝐿 −𝐿 𝐿 −𝐿 𝐿 −𝐿

ODD 0 0 2 𝐿
FUNCTIONS න 𝑓 𝑥 sin 𝑛𝑥 𝑑𝑥
𝐿 0

EVEN 2 𝐿 2 𝐿 0
FUNCCTION න 𝑓 𝑥 𝑑𝑥 න 𝑓 𝑥 cos 𝑛𝑥 𝑑𝑥
𝐿 0 𝐿 0
2. Piecewise continuous function
A function 𝑓(𝑡) is said to be piecewise continuous in an interval if
(1) 𝑓(𝑡) can be divided into finite number of subintervals.
(2) The limits of 𝑓(𝑡) at end points are finite.
3. Types of functions (Three types)

Even Odd

Symmetric about vertical axis Symmetric about the Origin

3. Types of functions (Three types)

Neither even nor odd

Has no Symmetry
 Even and Odd functions
A fnction f ( x ) even function if it is symmetric about y - axis A fnction f ( x) Odd function if it is symmetric about the origin
y

For example: For example:

cos x, x2 , x4 , x sin x, x, x3 , x5
 What is the definition of Fourier Series?

Now, Fourier Series can be defined as;

𝒂𝟎 ∞ 𝑛𝜋𝑡 ∞ 𝑛𝜋𝑡
𝑓 𝑡 = +෍ 𝒂𝒏 𝑐𝑜𝑠 +෍ 𝒃𝒏 𝑠𝑖𝑛
2 𝑛=1 𝐿 𝑛=1 𝐿

where

𝐿: is the half of the period

𝑎0 , 𝑎𝑛 and 𝑏𝑛 are the Fourier constants defined by

𝟏 𝑳 𝟏 𝑳 𝒏𝝅 𝒕 𝟏 𝑳 𝒏𝝅 𝒕
𝒂𝟎 = න 𝒇 𝒕 𝒅𝒕 𝒂𝒏 = න 𝒇 𝒕 𝒄𝒐𝒔 𝒅𝒕 𝒃𝒏 = න 𝒇 𝒕 𝒔𝒊𝒏 𝒅𝒕
𝑳 −𝑳 𝑳 −𝑳 𝑳 𝑳 −𝑳 𝑳
 Important Rules

𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 ∗ 𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 = 𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏

𝒐𝒅𝒅 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 ∗ 𝒐𝒅𝒅 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 = 𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏

𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 ∗ 𝒐𝒅𝒅 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 = 𝒐𝒅𝒅 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏

𝑳
𝑳
න 𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 = 𝟐 න 𝒆𝒗𝒆𝒏 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏
−𝑳
𝟎

𝑳
න 𝒐𝒅𝒅 𝒇𝒖𝒏𝒄𝒕𝒊𝒐𝒏 = 𝟎
−𝑳
 Calculating Fourier constants by type of function

𝒇(𝒕) is even 𝒇(𝒕) is odd 𝒇(𝒕) is neither even nor odd

𝟏 𝑳
𝒃𝒏 = 𝟎 𝒂𝟎 = න 𝒇(𝒕)𝒅𝒕
𝑳 −𝑳
𝟐 𝑳 𝒂𝟎 = 𝒂𝒏 = 𝟎
𝒂𝟎 = න 𝒇(𝒕)𝒅𝒕 𝟏 𝑳 𝒏𝝅 𝒕
𝑳 𝟎 𝒂𝒏 = න 𝒇(𝒕) 𝒄𝒐𝒔 𝒅𝒕
𝟐 𝑳 𝒏𝝅 𝒕 𝑳 −𝑳 𝑳
𝒃𝒏 = න 𝒇(𝒕) 𝒔𝒊𝒏 𝒅𝒕
𝑳 𝟎 𝑳
𝟐 𝑳 𝒏𝝅 𝒕
𝒂𝒏 = න 𝒇(𝒕) 𝒄𝒐𝒔 𝒅𝒕 𝟏 𝑳 𝒏𝝅 𝒕
𝑳 𝟎 𝑳 𝒃𝒏 = න 𝒇(𝒕) 𝒔𝒊𝒏 𝒅𝒕
𝑳 −𝑳 𝑳
 Steps to find the Fourier series

1. We plot the function during the base period.

2. We repeat the function more than once as required, or at least we repeat
once right and left.
3. From the graph we determine the type of the function and from the data we
calculate the half-period L.
4. By the type of function we calculate the constants corresponding to the case.
5. We substitute the constants into the general form of the series.
6. Sometimes the series is required to be used in a particular proof by
substituting an appropriate point.
Example 1: Find the Fourier series of the function f ( x) = x 2 on the interval −  , 


Solution: the general interval [ − L,L] L =   = =1
L ‫أوال نرسم الداله و نكررها‬
‫على اليمني و اليسار‬
‫ألن مفكوك فوريري يتعامل‬
‫مع الدوال التكرارية‬


a0
F(x) is even function , thus bn = 0 f ( x) = +  an cos nx
2 n=1
2L
where a =  f ( x ) dx
0
L0

2 2 2 x  2 2 3 3 2

=  f ( x ) dx =  x dx = = =
2

0 0   3  0  3 3
  
2 2 2  sin nx   cos nx   sin nx 
an =  f ( x) cosnx dx = 
2
x cos nx dx = ( x 2 )  − (2 x) −  + (2) − 
 0
 0
  n   n 
2
 n 3  0
2  cos n  4(−1)
n
= (2 )  =
  n
2
 n2
Example 2: Find the Fourier series of the function f ( x) = x on the interval 0,2 

Solution: the general interval [ 0 ,2 L] 2L = 2  L =    = 1

a0 
f ( x) = +  an cos nx + bn sin nx
From the graph, general function ( not even , not odd) 2 n =1
2 2 2
1 1 1  x2  1 2
1 2
a0 =
  0
f ( x) dx =
 
0
x dx =   = 2
  2 0
b =
n

 f (x ) sin nx dx =
0 
 x sin nx dx
0

2
1
2
1
2
1  − cos nx   − sin nx 
an =  f ( x) cosnx dx =  x cos nx dx = (x )   + (1)  
 0
 0
diff int   n   n
2
 0
2 x cos nx
1   sin nx   cos nx  + 1 cos 2n 
diff int
= ( x)  + (1)  sin nx =  −2 − 0
  n   n 2
 0 1
n  n 
x
+
sin nx
- cos nx
1  cos 2n 1 cos nx 1  −2  2 1
-
-
=  − 0 - =  = − n
  n2 n 2  n2
  n  n sin nx
0 - 2
1  ( −1) 2 n − 1 n
=  =0
 n2
Example (3)
−3 −2<𝑡 <0 1 1 1 𝜋
Find Fourier series for 𝑓 𝑡 = ቊ , period = 4, Then prove that 1 − + − +. . . =
3 0<𝑡<2 3 5 7 4

Solution
 from the graph we found the symmetry about the origen 𝑡ℎ𝑒𝑟𝑒 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑖𝑠 𝑜𝑑𝑑 𝐿 = 2

Then 𝑎0 = 𝑎𝑛 = 0 𝑎𝑛𝑑 𝑤𝑒 𝑔𝑒𝑡 𝒃𝒏 only

2
𝐿 2
2 𝑛𝜋𝑡 2 𝑛𝜋𝑡 2 𝑛𝜋𝑡 6
𝑏𝑛 = න 𝑓 𝑡 𝑠𝑖𝑛 𝑑𝑡 = න 3 𝑠𝑖𝑛 𝑑𝑡 = − 3 ∗ 𝑐𝑜𝑠 อ =− 𝑐𝑜𝑠 𝑛𝜋 − 1
𝐿 0 𝐿 2 0 2 𝑛𝜋 2 𝑛𝜋
0
 2
2 𝐿 𝑛𝜋𝑡 2 2 𝑛𝜋𝑡 2 𝑛𝜋𝑡 6
𝑏𝑛 = න 𝑓 𝑡 𝑠𝑖𝑛 𝑑𝑡 = න 3 𝑠𝑖𝑛 𝑑𝑡 = − 3 ∗ 𝑐𝑜𝑠 อ =− 𝑐𝑜𝑠 𝑛𝜋 − 1
𝐿 0 𝐿 2 0 2 𝑛𝜋 2 𝑛𝜋
0

Note that
𝑐𝑜𝑠 𝑛𝜋 = −1 𝑛 , 𝑐𝑜𝑠(2𝑛𝜋) = 1, 𝑠𝑖𝑛 𝑛𝜋 = 0, 𝑛 = 1,2. . .

6 0 𝐹𝑜𝑟 𝑛 𝑒𝑣𝑒𝑛
∴ 𝑏𝑛 = 1 − −1 𝑛 = ቐ 12
𝑛𝜋 𝐹𝑜𝑟 𝑛 𝑜𝑑𝑑
𝑛𝜋

Substitute in Fourier series form

∞ 𝟎 𝑭𝒐𝒓 𝒏 𝒆𝒗𝒆𝒏
𝟔 𝒏
𝒏𝝅𝒕 ∞ 𝟏𝟐 𝒏𝝅𝒕
𝒇 𝒕 =෍ 𝟏 − −𝟏 𝒔𝒊𝒏 = ቐ෍ 𝒔𝒊𝒏
𝒏=𝟏 𝒏𝝅 𝟐
𝒏=𝟏,𝟑,𝟓,... 𝒏𝝅 𝟐
 12 𝜋𝑡 12 3𝜋𝑡 12 5𝜋𝑡
𝑓 𝑡 = 𝑠𝑖𝑛 + 𝑠𝑖𝑛 + 𝑠𝑖𝑛 +. . .
𝜋 2 3𝜋 2 5𝜋 2
We take a common factor
12 𝜋𝑡 1 3𝜋𝑡 1 5𝜋𝑡
𝑓(𝑡) = 𝑠𝑖𝑛 + 𝑠𝑖𝑛 + 𝑠𝑖𝑛 +. . .
𝜋 2 3 2 5 2
From comparing the terms from the above equations, and we get
𝜋𝑡 3𝜋𝑡 5𝜋𝑡
𝑠𝑖𝑛 = 1, 𝑠𝑖𝑛 = −1, 𝑠𝑖𝑛 = 1, . . .
2 2 2
We get the suitable value for t and we found 𝑡 = 1
We substitute in Fourier series about t and we found
12 𝜋(1) 1 3𝜋(1) 1 5𝜋(1) 12 1 1
𝑓(1) = 𝑠𝑖𝑛 + 𝑠𝑖𝑛 + 𝑠𝑖𝑛 +. . . = 1 − + ...
𝜋 2 3 2 5 2 𝜋 3 5
 half interval Fourier expansion
In this case we give the function f (x ) defined on half interval [0,L] and we want to expand it to the full interval [-L,L]

Fourier cosine series Fourier sine series

means even function

means odd function
Example 3: Expand in Fourier sine series the function
f ( x) = x 2 , x  0, 

Solution: in this case, we extend the function f (x ) so that the new function defined over −  , 
is odd function as shown 
= = 1.
L

Odd function means a0 = an = 0


thus f ( x ) =  bn sin nx
n =1

 
2 2
where bn =
 
0
f ( x) sinnx dx =
 
0
x 2 sin nx dx Integrating by parts

2  cos nx   sin nx   cos nx 
bn = ( x 2 ) −  + (2 x) −  + ( 2 ) 
  n   n 2
  n 3
 0
2  2  (−1) n   (−1) n   1 
= ( ) −  + (2 )(0 ) + (2) 3  − (2) 3 
 n n n 
Example 4: Expand in Fourier cosine series the function

f (x ) = x , x  [0,  ]
Solution: in this case, we extend the function f (x ) so that the new function defined over −  , 
is even function as shown

even function means b =0n

thus

a0 
f ( x ) = +  an cos nx
2 n =1
2 L

a =  f (x ) dx
0
L 0

2 

2 2 x  2 2

=  x dx =   = 
 0 2 2 0
 = 
2 2
a =  f (x ) cos nx dx =  x cos nx dx Integrating by parts
n
 0  0

2  sin nx   cos nx 
= (x )   − (1)  − 
  n   n 2
 0

2  ( −1) 1 n

=  − 
 n 2
n 2