R20A_ADVC_UNIT-4_HANDOUT

FOURIER SERIES
 CONTENTS
UNIT-4
FOURIER SERIES
 Introduction to Fourier Series

 Periodic Functions

 Euler’s Formulae

 Definition of Fourier Series

 Fourier Series defined in various Intervals

 Half Range Fourier Series

 Important Formulae

 Problems on Fourier series

 FOURIER SERIES
Fourier Series is an infinite series representation of periodic function in terms of the
trigonometric sine and cosine functions.

Most of the single valued functions which occur in applied mathematics can be expressed in the
form of Fourier series, which is in terms of sines and cosines.

Fourier series is to be expressed in terms of periodic functions- sines and cosines.

Fourier series is a very powerful method to solve ordinary and partial differential equations,
particularly with periodic functions appearing as non-homogeneous terms.

We know that, Taylor’s series expansion is valid only for functions which are continuous and
differentiable. Fourier series is possible not only for continuous functions but also for periodic
functions, functions which are discontinuous in their values and derivatives. Further, because of
the periodic nature, Fourier series constructed for one period is valid for all values.

Periodic Functions

A function 𝑓(𝑥) is said to be periodic function with period 𝑇 > 0 if for all , 𝑓(𝑥 + 𝑇) = 𝑓(𝑥) , and 𝑇
is the least of such values.

Ex: 1) sin 𝑥 , cos 𝑥 are periodic functions with period 2𝜋.

2) tan 𝑥 , cot 𝑥 are periodic functions with period 𝜋.

Euler’s Formulae
The Fourier Series for the function 𝑓(𝑥) in the interval 𝐶 ≤ 𝑥 ≤ 𝐶 + 2𝜋 is given by

𝑎0 ∞
𝑓(𝑥) = + ∑𝑛=1(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2

1 𝐶+2𝜋
where 𝑎0 = 𝜋 ∫𝐶 𝑓(𝑥) 𝑑𝑥

1 𝐶+2𝜋
𝑎𝑛 = 𝜋 ∫𝐶 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝐶+2𝜋
𝑏𝑛 = 𝜋 ∫𝐶 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are known as Euler’s Formulae.

CONDITIONS FOR FOURIER EXPANSION (𝐷𝑖𝑟𝑐ℎ𝑙𝑒𝑡 𝐶𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑠)

A function 𝑓(𝑥) defined in [0, 2𝜋] has a valid Fourier series expansion of the form

𝑎0 ∞
+ ∑𝑛=0(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2

Where 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are constants, provided

1) 𝑓(𝑥) is well defined and single-valued, except possibly at a finite number of point in the
interval [0, 2𝜋].
2) 𝑓(𝑥) has finite number of finite discontinuities in the interval in [0, 2𝜋].
3) 𝑓(𝑥) has finite number of finite maxima and minima.

Note: The above conditions are valid for the function defined in the Intervals [– 𝜋, 𝜋][0, 2𝑙], [−𝑙, 𝑙] .

{1, cos 1𝑥, cos 2𝑥, cos 3𝑥, … , cos 𝑛𝑥, … , sin 1𝑥, sin 2𝑥 , sin 3𝑥, … , sin 𝑛𝑥, … }
Consider any two, All these have a common period 2𝜋. Here 1 =cos 0𝑥
𝜋𝑥 2𝜋𝑥 3𝜋𝑥 𝑛𝜋𝑥 𝜋𝑥 2𝜋𝑥 3𝜋𝑥 𝑛𝜋𝑥
{1, cos , cos , cos , … , cos , … , sin , sin , sin , … , sin ,… }
𝑙 𝑙 𝑙 𝑙 𝑙 𝑙 𝑙 𝑙

All these have a common period 2𝑙.

These are called complete set of orthogonal functions.

Definition of Fourier series

Let 𝑓(𝑥) be a function defined in [0, 2𝜋]. Let 𝑓(𝑥 + 2𝜋) = 𝑓(𝑥) ∀ 𝑥, then the Fourier Series of
𝑎0 ∞
𝑓(𝑥) is given by 𝑓(𝑥) = + ∑𝑛=1(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
1 2𝜋
where 𝑎0 = 𝜋 ∫0 𝑓(𝑥) 𝑑𝑥
1 2𝜋
𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 2𝜋
𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are called as Fourier coefficients of 𝑓(𝑥) in [0, 2𝜋].

Let 𝑓(𝑥) be a function defined in [−𝜋, 𝜋]. Let 𝑓(𝑥 + 2𝜋) = 𝑓(𝑥) ∀ 𝑥, then the Fourier Series of
𝑎0 ∞
𝑓(𝑥) is given by 𝑓(𝑥) = + ∑𝑛=1(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
1 𝜋
where 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥

1 𝜋
𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝜋
𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are called as Fourier coefficients of 𝑓(𝑥) in [−𝜋, 𝜋].

 Let 𝑓(𝑥) be a function defined in [0, 2𝑙]. Let 𝑓(𝑥 + 2𝑙) = 𝑓(𝑥) ∀ 𝑥, then the Fourier Series of
𝑎0 𝑛𝜋𝑥 𝑛𝜋𝑥
𝑓(𝑥) is given by 𝑓(𝑥) = + ∑∞
𝑛=1 (𝑎𝑛 cos + 𝑏𝑛 sin )
2 𝑙 𝑙
1 2𝑙
where 𝑎0 = ∫0 𝑓(𝑥) 𝑑𝑥
𝑙
1 2𝑙 𝑛𝜋𝑥
𝑎𝑛 = ∫0 𝑓(𝑥) cos 𝑑𝑥
𝑙 𝑙

1 2𝑙 𝑛𝜋𝑥
𝑏𝑛 = ∫0 𝑓(𝑥) sin 𝑑𝑥
𝑙 𝑙

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are called as Fourier coefficients of 𝑓(𝑥) in [0, 2𝑙].

Let 𝑓(𝑥) be a function defined in [−𝑙, 𝑙]. Let 𝑓(𝑥 + 2𝜋) = 𝑓(𝑥) ∀ 𝑥, then the Fourier Series of
𝑎0 ∞ 𝑛𝜋𝑥 𝑛𝜋𝑥
𝑓(𝑥) is given by 𝑓(𝑥) = + ∑𝑛=1 (𝑎𝑛 cos + 𝑏𝑛 sin )
2 𝑙 𝑙

1 𝑙
where 𝑎0 = ∫−𝑙 𝑓(𝑥) 𝑑𝑥
𝑙
1 𝑙 𝑛𝜋𝑥
𝑎𝑛 = ∫−𝑙 𝑓(𝑥) cos 𝑑𝑥
𝑙 𝑙

1 𝑙 𝑛𝜋𝑥
𝑏𝑛 = ∫−𝑙 𝑓(𝑥) sin 𝑑𝑥
𝑙 𝑙

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are called as Fourier coefficients of 𝑓(𝑥) in [−𝑙, 𝑙].

FOURIER SERIES FOR EVEN AND ODD FUNCTIONS

We know that if 𝑓(𝑥) be a function defined in [−𝜋, 𝜋]. Let 𝑓(𝑥 + 2𝜋) = 𝑓(𝑥) ∀ 𝑥, then the Fourier
𝑎0 ∞
Series of 𝑓(𝑥) is given by 𝑓(𝑥) = + ∑𝑛=1(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
1 𝜋
where 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥
1 𝜋
𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝜋
𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

These values 𝑎0 , 𝑎𝑛 , 𝑏𝑛 are called as Fourier coefficients of 𝑓(𝑥) in [−𝜋, 𝜋].

Case (i): When 𝑓(𝑥) is an even function

1 𝜋 2 𝜋
then, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥 = 𝜋 ∫0 𝑓(𝑥) 𝑑𝑥

Since cos 𝑛𝑥 is an even function, 𝑓(𝑥) is an even function ⟹ Product of two even functions is even

1 𝜋 2 𝜋
∴ 𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 = ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 (∵ 𝐼𝑛𝑡𝑒𝑔𝑟𝑎𝑛𝑑 𝑖𝑠 𝑒𝑣𝑒𝑛)
𝜋

Now, sin 𝑛𝑥 is an odd function, 𝑓(𝑥) is an even function ⟹ Product of odd and even is odd
 1 𝜋
∴ 𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥 = 0 (∵ 𝐼𝑛𝑡𝑒𝑔𝑟𝑎𝑛𝑑 𝑖𝑠 𝑜𝑑𝑑)

Thus, if a function 𝑓(𝑥) is even in [−𝜋, 𝜋], its Fourier series expansion contains only cosine terms.

𝑎 ∞
Hence Fourier Series is given by 𝑓(𝑥) = 20 + ∑𝑛=1 𝑎𝑛 cos 𝑛𝑥

2 𝜋
where 𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 , 𝑛 = 0, 1, 2, 3, …

Case (ii): When 𝑓(𝑥) is an Odd Function

1 𝜋
then, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥 = 0

Since cos 𝑛𝑥 is an even function, 𝑓(𝑥) is an odd function ⟹ Product of even and odd is even

1 𝜋
∴ 𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 = 0 (∵ 𝐼𝑛𝑡𝑒𝑔𝑟𝑎𝑛𝑑 𝑖𝑠 𝑜𝑑𝑑)

Now, sin 𝑛𝑥 is an odd function, 𝑓(𝑥) is an odd function ⟹ Product of two odd functions is even

1 𝜋 2 𝜋
∴ 𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥 = 𝜋 ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥 (∵ 𝐼𝑛𝑡𝑒𝑔𝑟𝑎𝑛𝑑 𝑖𝑠 𝑒𝑣𝑒𝑛)

Thus, if a function 𝑓(𝑥) is Odd in [−𝜋, 𝜋], its Fourier series expansion contains only sine terms.

Hence, if 𝑓(𝑥) is odd function defined in [−𝜋, 𝜋], 𝑓(𝑥) can be expanded as a series of the form

𝑓(𝑥) = ∑∞
𝑛=1 𝑏𝑛 sin 𝑛𝑥

2 𝜋
where, 𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

HALF RANGE FOURIER SERIES

Half Range Fourier Sine Series defined in [0, 𝜋]: The Fourier half range sine series in [0, 𝜋] is given
by
∞

𝑓(𝑥) = ∑ 𝑏𝑛 𝑠𝑖𝑛 𝑛𝑥,

𝑛=1
2 𝜋
where, 𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) 𝑠𝑖𝑛 𝑛𝑥 𝑑𝑥

This is Similar to the Fourier series defined for odd function in [– 𝜋, 𝜋]

Half Range Fourier Cosine Series defined in [0, 𝜋]: The Fourier half range Cosine series in [0, 𝜋] is
given by
 ∞
𝑎0
𝑓(𝑥) = + ∑ 𝑎𝑛 𝑐𝑜𝑠 𝑛𝑥
2
𝑛=1
2 𝜋
where, 𝑎0 = ∫ 𝑓(𝑥) 𝑑𝑥
𝜋 0
2 𝜋
𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) 𝑐𝑜𝑠 𝑛𝑥 𝑑𝑥

This is Similar to the Fourier series defined for even function in [– 𝜋, 𝜋]

Half Range Fourier Sine Series defined in [0, 𝑙]: The Fourier half range sine series in [0, 𝜋] is given
by
∞
𝑛𝜋𝑥
𝑓(𝑥) = ∑ 𝑏𝑛 𝑠𝑖𝑛
𝑙
𝑛=1
2 𝑙 𝑛𝜋𝑥
where, 𝑏𝑛 = ∫0 𝑓(𝑥) 𝑠𝑖𝑛 𝑑𝑥
𝑙 𝑙

This is Similar to the Fourier series defined for odd function in [– 𝑙, 𝑙]

Half Range Fourier Cosine Series defined in [0, 𝑙]: The Fourier half range Cosine series in [0, 𝑙] is
given by
∞
𝑎0 𝑛𝜋𝑥
𝑓(𝑥) = + ∑ 𝑎𝑛 𝑐𝑜𝑠
2 𝑙
𝑛=1
2 𝑙
where, 𝑎0 = 𝑙 ∫0 𝑓(𝑥) 𝑑𝑥
2 𝑙 𝑛𝜋𝑥
𝑎𝑛 = ∫0 𝑓(𝑥) 𝑐𝑜𝑠 𝑑𝑥
𝑙 𝑙

This is Similar to the Fourier series defined for even function in [– 𝑙, 𝑙]

Important Formulae

𝑎
2𝑎 2 ∫0 𝑓(𝑥) 𝑑𝑥 𝑖𝑓 𝑓(2𝑎 − 𝑥) = 𝑓(𝑥)
∫0 𝑓(𝑥) 𝑑𝑥 = {
0 𝑖𝑓 𝑓(2𝑎 − 𝑥) = −𝑓(𝑥)
𝑎
𝑎 2 ∫0 𝑓(𝑥) 𝑑𝑥 𝑖𝑓 𝑓 𝑖𝑠 𝑒𝑣𝑒𝑛
∫−𝑎 𝑓(𝑥) 𝑑𝑥 = {
0 𝑖𝑓 𝑓 𝑖𝑠 𝑜𝑑𝑑
Here Even function means: If 𝑓(−𝑥) = 𝑓(𝑥), then 𝑓(𝑥) is called as even function
Odd function means : If 𝑓(−𝑥) = −𝑓(𝑥), then 𝑓(𝑥) is called as odd function.
𝑏 𝑎
∫𝑎 𝑓(𝑥) 𝑑𝑥 = − ∫𝑏 𝑓(𝑥) 𝑑𝑥
𝑏 𝑐 𝑏
∫𝑎 𝑓(𝑥) 𝑑𝑥 = ∫𝑎 𝑓(𝑥) 𝑑𝑥 + ∫𝑐 𝑓(𝑥) 𝑑𝑥 ; 𝑎 < 𝑐 < 𝑏
𝑎 𝑎 𝑏 𝑏
∫0 𝑓(𝑥) 𝑑𝑥 = ∫0 𝑓(𝑎 − 𝑥) 𝑑𝑥 , Also ∫𝑎 𝑓(𝑥) 𝑑𝑥 = ∫𝑎 𝑓(𝑎 + 𝑏 − 𝑥) 𝑑𝑥
Problems on Fourier Series
1) Find the Fourier series to represent 𝒇(𝒙) = 𝒙𝟐 in the interval (𝟎, 𝟐𝝅).

Sol: We know that, the Fourier series of 𝑓(𝑥) defined in the interval (0, 2𝜋) is given by
∞
𝑎0
𝑓(𝑥) = + ∑(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
𝑛=1

1 2𝜋
where, 𝑎0 = 𝜋 ∫0 𝑓(𝑥) 𝑑𝑥

1 2𝜋
𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 2𝜋
𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

Here, 𝑓(𝑥) = 𝑥 2
1 2𝜋 1 2𝜋
Now, 𝑎0 = 𝜋 ∫0 𝑓(𝑥) 𝑑𝑥 = 𝜋 ∫0 𝑥 2 𝑑𝑥

2𝜋
1 𝑥3 1 8
= 𝜋[3] = 3𝜋 [(2𝜋)3 − 0] = 3 𝜋 2
0

8
⟹ 𝑎0 = 3 𝜋 2

1 2𝜋 1 2𝜋
Again, 𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 = 𝜋 ∫0 𝑥
⏟2 cos
⏟ 𝑛𝑥 𝑑𝑥
𝑢 𝑣

1 𝑑
= 𝜋 [𝑥 2 ∫ cos 𝑛𝑥 𝑑𝑥 − {∫ 𝑑𝑥 (𝑥 2 )(∫ cos 𝑛𝑥 𝑑𝑥)𝑑𝑥}]

𝑑𝑢
[∵ ∫ 𝑢𝑣 𝑑𝑥 = 𝑢 ∫ 𝑣 𝑑𝑥 − {∫ 𝑑𝑥 . (∫ 𝑣 𝑑𝑥)𝑑𝑥}]

1 sin 𝑛𝑥 sin 𝑛𝑥 2𝜋
= 𝜋 [𝑥 2 ( ) − {∫ 2𝑥 ( ) 𝑑𝑥}]
𝑛 𝑛 0

2𝜋
1 sin 𝑛𝑥 2
= [𝑥 2 ( 𝑛 ) − 𝑛 {∫ ⏟
𝑥 sin
⏟ 𝑛𝑥 𝑑𝑥}]
𝜋
𝑢 𝑣 0

1 sin 𝑛𝑥 2 cos 𝑛𝑥 cos 𝑛𝑥 2𝜋

= 𝜋 [𝑥 2 ( ) − 𝑛 (−𝑥 + ∫ 1. 𝑑𝑥)]
𝑛 𝑛 𝑛 0

1 sin 𝑛𝑥 2 cos 𝑛𝑥 1 2𝜋
= 𝜋 [𝑥 2 ( ) − 𝑛 (−𝑥 + 𝑛 ∫ cos 𝑛𝑥 𝑑𝑥)]
𝑛 𝑛 0

1 sin 𝑛𝑥 2 cos 𝑛𝑥 1 sin 𝑛𝑥 2𝜋

= 𝜋 [𝑥 2 ( ) − 𝑛 (−𝑥 +𝑛 )]
𝑛 𝑛 𝑛 0

1 sin 𝑛𝑥 2 2 2𝜋
= 𝜋 [𝑥 2 ( ) + 𝑛2 𝑥 cos 𝑛𝑥 − 𝑛3 sin 𝑛𝑥]
𝑛 0

4 cos 2𝑛𝜋 = 1
= 𝑛2 [∵ ]
sin 2𝑛𝜋 = 0
 4
⟹ 𝑎𝑛 = 𝑛 2

1 2𝜋 1 2𝜋
Again, 𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥 = 𝜋 ∫0 𝑥
⏟2 sin
⏟ 𝑛𝑥 𝑑𝑥
𝑢 𝑣

1 𝑑
= 𝜋 [𝑥 2 ∫ sin 𝑛𝑥 𝑑𝑥 − {∫ 𝑑𝑥 (𝑥 2 )(∫ sin 𝑛𝑥 𝑑𝑥)𝑑𝑥}]

𝑑𝑢
[∵ ∫ 𝑢𝑣 𝑑𝑥 = 𝑢 ∫ 𝑣 𝑑𝑥 − {∫ 𝑑𝑥 . (∫ 𝑣 𝑑𝑥)𝑑𝑥}]

1 cos 𝑛𝑥 cos 𝑛𝑥 2𝜋
= 𝜋 [𝑥 2 (− ) − {∫ 2𝑥 (− ) 𝑑𝑥}]
𝑛 𝑛 0

2𝜋
1 cos 𝑛𝑥 2
= [−𝑥 2 ( 𝑛 ) + 𝑛 {∫ ⏟
𝑥 cos
⏟ 𝑛𝑥 𝑑𝑥}]
𝜋
𝑢 𝑣 0

1 cos 𝑛𝑥 2 sin 𝑛𝑥 sin 𝑛𝑥 2𝜋

= 𝜋 [−𝑥 2 ( ) + 𝑛 (𝑥 + ∫ 1. 𝑑𝑥)]
𝑛 𝑛 𝑛 0

1 cos 𝑛𝑥 2 sin 𝑛𝑥 1 2𝜋
= 𝜋 [−𝑥 2 ( ) + 𝑛 (𝑥 + 𝑛 ∫ sin 𝑛𝑥 𝑑𝑥)]
𝑛 𝑛 0

1 cos 𝑛𝑥 2 sin 𝑛𝑥 1 cos 𝑛𝑥 2𝜋

= 𝜋 [−𝑥 2 ( ) + 𝑛 (𝑥 +𝑛 )]
𝑛 𝑛 𝑛 0

1 cos 𝑛𝑥 2 2 2𝜋
= 𝜋 [−𝑥 2 ( ) + 𝑛2 𝑥 sin 𝑛𝑥 + 𝑛3 cos 𝑛𝑥]
𝑛 0

4𝜋 cos 2𝑛𝜋 = 1
=− [∵ ]
𝑛 sin 2𝑛𝜋 = 0
4𝜋
⟹ 𝑏𝑛 = − 𝑛

∞
𝑎0
∴ 𝑓(𝑥) = + ∑(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
𝑛=1

8𝜋2
4 4𝜋
2
∴ 𝑓(𝑥) = 𝑥 = 3
+ ∑∞
𝑛=1 (𝑛2 cos 𝑛𝑥 − sin 𝑛𝑥)
2 𝑛

4𝜋 2 4 4𝜋
⟹ 𝑥2 = + ∑∞
𝑛=1 (𝑛2 cos 𝑛𝑥 − sin 𝑛𝑥)
3 𝑛

−𝝅 ; −𝝅 < 𝑥 < 0
2) Find the Fourier series of the periodic function defined as 𝒇(𝒙) = {
𝒙 ; 𝟎<𝑥<𝜋
𝟏 𝟏 𝟏 𝝅𝟐
Hence deduce that + 𝟑𝟐 + 𝟓𝟐 + … =
𝟏𝟐 𝟖

Sol: We know that, the Fourier series of 𝑓(𝑥) defined in the interval (−𝜋, 𝜋) is given by
 ∞
𝑎0
𝑓(𝑥) = + ∑(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
𝑛=1

1 𝜋
where, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥

1 𝜋
𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝜋
𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

−𝜋 ; −𝜋 < 𝑥 < 0
Here, 𝑓(𝑥) = {
𝑥 ; 0<𝑥<𝜋
1 𝜋 1 0 𝜋
Now, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥 = 𝜋 [∫−𝜋 𝑓(𝑥) 𝑑𝑥 + ∫0 𝑓(𝑥) 𝑑𝑥]

1 0 𝜋
= 𝜋 [∫−𝜋(−𝜋) 𝑑𝑥 + ∫0 𝑥 𝑑𝑥]

1 0 𝜋
= 𝜋 [(−𝜋) ∫−𝜋 𝑑𝑥 + ∫0 𝑥 𝑑𝑥]

𝜋
1 𝑥2 1 𝜋2
= 𝜋 [(−𝜋)[𝑥]0−𝜋 + [ 2 ] ] = 𝜋 [(−𝜋)(𝜋) + 2
]
0

1 𝜋2 𝜋
= 𝜋 [−𝜋 2 + ]= −2
2

𝜋
⟹ 𝑎0 = −
2
1 𝜋
Also, 𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 0 𝜋
= 𝜋 [∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥 + ∫0 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥]

1 0 𝜋
= 𝜋 [∫−𝜋(−𝜋) cos 𝑛𝑥 𝑑𝑥 + ∫0 𝑥 cos 𝑛𝑥 𝑑𝑥]

1 0 𝜋
= 𝜋 [−𝜋 ∫−𝜋(cos 𝑛𝑥) 𝑑𝑥 + ∫0 𝑥 cos 𝑛𝑥 𝑑𝑥]

1 sin 𝑛𝑥 0 sin 𝑛𝑥 sin 𝑛𝑥 𝜋

= 𝜋 [−𝜋 ( ) + {𝑥 ( ) − ∫1( ) 𝑑𝑥 } ]
𝑛 −𝜋 𝑛 𝑛 0

1 𝜋 𝑥 sin 𝑛𝑥 1 𝜋
= 𝜋 [− 𝑛 (sin 𝑛𝑥)0−𝜋 + { − 𝑛 ∫ sin 𝑛𝑥 𝑑𝑥 } ]
𝑛 0

1 𝜋 𝑥 sin 𝑛𝑥 1 (− cos 𝑛𝑥) 𝜋

= 𝜋 [− 𝑛 (sin 𝑛𝑥)0−𝜋 + { −𝑛 } ]
𝑛 𝑛 0

1 𝜋 𝑥 sin 𝑛𝑥 1 𝜋
= 𝜋 [− 𝑛 (sin 𝑛𝑥)0−𝜋 + { 𝑛
+ 𝑛2 cos 𝑛𝑥} ]
0

1 𝜋 𝜋 sin 𝑛𝜋 1 1
= 𝜋 [− 𝑛 [0 − sin(−𝑛𝜋)] + {( + 𝑛2 cos 𝑛𝜋) − (0 + 𝑛2 cos 𝑛0)}]
𝑛

1 𝜋 𝜋 sin 𝑛𝜋 1 1 cos(−𝜃) = cos 𝜃

= 𝜋 [− 𝑛 sin 𝑛𝜋 + {( + 𝑛2 cos 𝑛𝜋) − 𝑛2 .1}] [∵ ]
𝑛 sin(−𝜃) = sin 𝜃
 sin 𝑛𝜋 = 0
1 𝜋 𝜋(0) 1 1
⟹ 𝑎𝑛 = 𝜋 [− 𝑛 (0) + {( + 𝑛2 (−1)𝑛 ) − 𝑛2 }] [∵ & ]
𝑛
cos 𝑛𝜋 = (−1)𝑛
1 (−1)𝑛 1 1
= 𝜋[ + 𝑛2 ] = 𝜋𝑛2 [(−1)𝑛 − 1]
𝑛2

1
⟹ 𝑎𝑛 = 𝜋𝑛2 [(−1)𝑛 − 1]

1 𝜋
Again, 𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

1 0 𝜋
= 𝜋 [∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥 + ∫0 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥]

1 0 𝜋
= 𝜋 [∫−𝜋(−𝜋) sin 𝑛𝑥 𝑑𝑥 + ∫0 𝑥 sin 𝑛𝑥 𝑑𝑥]

1 0 𝜋
= 𝜋 [−𝜋 ∫−𝜋(sin 𝑛𝑥) 𝑑𝑥 + ∫0 𝑥 sin 𝑛𝑥 𝑑𝑥]

1 cos 𝑛𝑥 0 cos 𝑛𝑥 cos 𝑛𝑥 𝜋

= 𝜋 [−𝜋 (− ) + {𝑥 (− ) − ∫ 1 (− ) 𝑑𝑥 } ]
𝑛 −𝜋 𝑛 𝑛 0

1 𝜋 𝑥 cos 𝑛𝑥 1 𝜋
= 𝜋 [𝑛 (cos 𝑛𝑥)0−𝜋 + {− + 𝑛 ∫ cos 𝑛𝑥 𝑑𝑥 } ]
𝑛 0

1 𝜋 𝑥 cos 𝑛𝑥 1 (sin 𝑛𝑥) 𝜋

= 𝜋 [𝑛 (cos 𝑛𝑥)0−𝜋 + {− +𝑛 } ]
𝑛 𝑛 0

1 𝜋 𝑥 cos 𝑛𝑥 1 𝜋
= 𝜋 [𝑛 (cos 𝑛𝑥)0−𝜋 + {− + 𝑛2 sin 𝑛𝑥} ]
𝑛 0

1 𝜋 𝜋 cos 𝑛𝜋 1 1
= 𝜋 [𝑛 [1 − cos(−𝑛𝜋)] + {(− + 𝑛2 sin 𝑛𝜋) − (−0 + 𝑛2 sin 𝑛0)}]
𝑛

1 𝜋 𝜋 cos 𝑛𝜋 1
= 𝜋 [𝑛 [1 − cos 𝑛𝜋] − ] = 𝑛 (1 − 2 cos 𝑛𝜋)
𝑛

sin 𝑛𝜋 = 0
cos(−𝜃) = cos 𝜃
[∵ & & ]
sin(−𝜃) = sin 𝜃 cos 𝑛𝜋 = (−1)𝑛

1
⟹ 𝑏𝑛 = 𝑛 (1 − 2 cos 𝑛𝜋)

Hence, the Fourier series for given 𝑓(𝑥) is given by

𝜋
− 1 1
𝑓(𝑥) = 2
+ ∑∞ 𝑛
𝑛=1 (𝜋𝑛2 [(−1) − 1] cos 𝑛𝑥 + 𝑛 (1 − 2 cos 𝑛𝜋) sin 𝑛𝑥)
2

𝜋 1 1
⟹ 𝑓(𝑥) = − 4 + ∑∞ 𝑛
𝑛=1 (𝜋𝑛2 [(−1) − 1] cos 𝑛𝑥 + 𝑛 (1 − 2 cos 𝑛𝜋) sin 𝑛𝑥)

𝜋 2 cos 3𝑥 cos 5𝑥 sin 2𝑥 sin 3𝑥 sin 4𝑥

⟹ 𝑓(𝑥) = − 4 − 𝜋 (cos 𝑥 + 32
+ 52
+ … ) + (3 sin 𝑥 − 2
+ 3
− 4
+ …)

Deduction: Put 𝑥 = 0 in the above function (𝑥) , we get

𝜋 2 1 1
𝑓(0) = − − (1 + 2 + 2 + … )
4 𝜋 3 5
 𝑓(0 − 0) = −𝜋
Since, 𝑓(𝑥) is discontinuous at 𝑥 = 0,
𝑓(0 + 0) = 0

1
⟹ 𝑓(0) = [𝑓(0 − 0) + 𝑓(0 + 0)]
2
1 𝜋
⟹ 𝑓(0) = (−𝜋) = −
2 2
𝜋 𝜋 2 1 1
Hence, 𝑓(0) = − 2 = − 4 − 𝜋 (1 + 32 + 52 + … )

1 1 1 𝜋2
⟹ + + + … =
12 32 52 8
Hence the result.

3) Expand the function 𝒇(𝒙) = 𝒙𝟐 as Fourier series in [– 𝝅, 𝝅].

𝟏 𝟏 𝟏 𝟏 𝝅𝟐
Hence deduce that + 𝟐𝟐 + 𝟑𝟐 + 𝟒𝟐 + … =
𝟏𝟐 𝟔

Sol: We know that, the Fourier series of 𝑓(𝑥) defined in the interval (−𝜋, 𝜋) is given by
∞
𝑎0
𝑓(𝑥) = + ∑(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)
2
𝑛=1

1 𝜋
where, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥

1 𝜋
𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝜋
𝑏𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

Here, 𝑓(𝑥) = 𝑥 2
1 𝜋
Now, 𝑎0 = 𝜋 ∫−𝜋 𝑓(𝑥) 𝑑𝑥

1 𝜋 2 𝜋
= 𝜋 ∫−𝜋 𝑥 2 𝑑𝑥 = 𝜋 ∫0 𝑥 2 𝑑𝑥

𝜋
2 𝑥3 2𝜋 2
= 𝜋[3] =
0 3

2𝜋 2
⟹ 𝑎0 = 3

1 𝜋
Again, 𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) cos 𝑛𝑥 𝑑𝑥

1 𝜋
= 𝜋 ∫−𝜋 𝑥 2 cos 𝑛𝑥 𝑑𝑥

2 𝜋 𝑎 𝑎
= 𝜋 ∫0 𝑥 2 cos 𝑛𝑥 𝑑𝑥 [∵ 𝑓(𝑥) 𝑖𝑠 𝑒𝑣𝑒𝑛 ⟹ ∫−𝑎 𝑓(𝑥)𝑑𝑥 = 2 ∫0 𝑓(𝑥)𝑑𝑥]
 𝜋
2 𝑥 2 sin 𝑛𝑥 2𝑥 cos 𝑛𝑥 2𝑥 𝑠𝑖𝑛 𝑛𝑥 4
= 𝜋[ + − ] = 𝑛2 (−1)𝑛
𝑛 𝑛2 𝑛3 0

4
⟹ 𝑎𝑛 = 𝑛2 (−1)𝑛

1 𝜋
Again, 𝑎𝑛 = 𝜋 ∫−𝜋 𝑓(𝑥) sin 𝑛𝑥 𝑑𝑥

1 𝜋
= 𝜋 ∫−𝜋 𝑥 2 sin 𝑛𝑥 𝑑𝑥

𝑎
=0 [∵ 𝑓(𝑥) 𝑖𝑠 𝑜𝑑𝑑 ⟹ ∫−𝑎 𝑓(𝑥)𝑑𝑥 = 0]

Hence, the Fourier series for given 𝑓(𝑥) is given by

2𝜋2
( ) 4
3
𝑓(𝑥) = 𝑥 = 2
+ ∑∞ 𝑛
𝑛=1 𝑛2 (−1) cos 𝑛𝑥
2

𝜋2 cos 2𝑥 cos 3𝑥 cos 4𝑥

⟹ 𝑥2 = + 4 (− cos 𝑥 + − + − …)
3 22 32 42

Deduction: Put 𝑥 = 𝜋 in the above equation, we get

𝜋2 cos 2𝜋 cos 3𝜋 cos 4𝜋
⟹ 𝜋2 = + 4 (− cos 𝜋 + − + − …)
3 22 32 42

𝜋2 1 1 1
⟹ 𝜋2 − = 4 (1 + 22 + 32 + 42 + … )
3

2𝜋 2 1 1 1
⟹ = 4 (1 + + + + …)
3 22 32 42

𝜋2 1 1 1
⟹ = 1 + 22 + 32 + 42 + …
6

∗∗∗