Signals & Systems

Fourier Series

1
 Contents
• Transform
• Basis functions
• Orthogonality
• Dirichlet conditions for existence of FS
• Trigonometric Fourier series
• Exponential Fourier Series
• Symmetry Properties
• Amplitude and phase spectrum,
• Gibbs phenomenon.

2
 Session Objectives
• Understand, and resolve the signals in
frequency domain using Fourier series

• Understand Representation of a periodic

CT Signals using basis signals like
sinusoids or complex exponentials.

3
 Session Outcomes
• Analyze and resolve the signals in
frequency domain using Fourier series

• Plot Amplitude and Phase Spectrum of

the various signals.

4
 Recap of Unit 1 and 2
• Any Signal can be represented as a linear
combination of some basis signals.

• Representations of CT signals as a linear combination

of shifted and scaled impulses

• Convolution: Output of any LTI System can be

represented as a linear combination of shifted and
scaled impulse responses.

5
What is a Transform and Why do We Need ?
 Transform: A mathematical operation that takes a
function or sequence and maps it into another one
 Why we need transform?
– To get additional /hidden information about the original
function, which may not be available /obvious otherwise
– The transform of an equation may be easier to solve than
the original equation
– The transform of a function/sequence may require less
storage, hence provide data compression / reduction
– An operation may be easier to apply on the transformed
function, rather than the original function (recall
convolution)
General Scheme using Transforms

Equation Solution
Problem Result
of the problem of the equation

Transformation
= HARD
Inverse
transformation
Transformed Solution of the
equation transformed equation
= EASY
Properties of Transforms

 Most useful properties are:

– Linear: we can pull out constants, and apply superposition
T (f   g)  T ( f )  T (g)
– One-to-one: different functions have different transforms
– Invertible: (kind of – sort of the undo button…)
f T F T-1 f

 Continuous transform: map functions to functions

 Discrete transform: map sequences to sequences
 How Does a Transform Look Like?

 Complex function representation through simple

building blocks
Complex Function   weight   Basis
i Functi
i

f   Fi K i j Fi   K ij f j (
i j

f (x)   K (x,  )F (

F ( ) 

 Compressed representation through using only a few

blocks (called basis functions / kernels)
 Sinusoids as building blocks: Fourier transform
– Frequency domain representation of the function
 Basis Function
• Every continuous function in the function
space can be represented as a linear
combination of basis functions.

10
 What Transforms are Available?

 Fourier series

 Continuous Fourier transform

1
F ()   f (t)e  jt dt
 Laplace transform
1
F (s)   f (t)e dtst

 Discrete Fourier transform

N 1  j 2 k
n N 1 j 2n
k
F[k]   f [n]e  F[ k]e
1
N
f[n]  N

n0 2 k 0
 Z-transform
N 1
1 N 1
F[z]   H.R.
f[n]e  zn
f[n]   F[z]e zn
n0
Pourreza 2 z0
Revision: Complex exponential Notation
• Euler’s formula eix  cos(x)  i sin( x)

Phasor notation:

x  iy  z ei

where z  x 2
 y 2

 z z
 ( x  iy )( x  iy )

  tan 1  y 
and  
 x 

12
 Euler’s formula
• Taylor series expansions
x2 x3 x4
e x  1 x    ...
2! 3! 4!
• Even function ( f(x) = f(-x) )
x2 x4 x6 x8
cos(x) 1    ...
2! 4! 6! 8!
• Odd function ( f(x) = -f(-x) )3
x x5 x7 x9
sin( x)  x     ...
3! 5! 7! 9!
x2 ix 3 x 4
ix 5 x 6
ix 7
eix  1 ix       ...
2! 3! 4! 5! 6! 7!
 cos(x)  i sin( x)
13
Sine and cosine waveform

14
 Standard values

21 January 2021 15
 Orthogonal Functions
• Call a set of functions {k} orthogonal
on an interval a < t < b if it satisfies

0 m n

b
 m (t ) n (t)dt  
a
rn m n

16
 Orthogonal set of Sinusoidal Functions

Define 0=2/T.
T/2 T /2
T /2
cos(m0t)dt  0, m 0 
T /2
sin( m 0t)dt  0, m 0

T /2  0 m n
T /2 cos(m0t) cos(n0t)dt  T / 2 m  n We now prove this one

T /2  0
T /2
sin( m 0t) sin( nt

T /2

21 January
T /2 2021
sin( m 0 t) cos(n 0 t)dt  0, for all m and n 17
 Proof
1
cos cos  [cos(  )  cos( )]
2


T /2

T /2
cos(m0t) cos(n 0t)dt mn
1 T /2 1 T /2
  cos[(m  n)0 t]dt  cos[(m  n)0t]dt
2 T /2 2 T /2
1 1 1 1
 sin[( m  n)0t]
T /2

2 (m n)
1 1 1 1
 2sin[( m  n)] 
2 (m n)
0

0
0
21 January 2021 18
 Proof
1
cos   [1
2

T /2

T /2
cos(m0t) cos(n 0t)dt m=n
1 T /2
  cos (m0t)dt  T /2 [1 cos 2m0t]dt
T /2
2
T /2 2
T /2 T /2
1 1
 t  sin 2m 0t]
2 T /2 4m0 T /2

0
T /2  0 m n

T
T /2 cos(m0t) cos(n0t)dt  T / 2 m  n
2
19
Orthogonal set of Sinusoidal Functions
1,

cos0t,cos20t,cos30t,
sin t,sin 2t,sin 3t,
 0 0 0

Define 0=2/T.
T /2 T /2

T /2
cos(m0t)dt  0, m 0 
T /2
sin( m 0t)dt  0, m 0

T /2  0 m n
T /2 cos(m0t) cos(n0 t)dt  T / 2 m  n
T /2  0 m n
T /2 sin( m0t) sin( n0t)dt  T / 2 m  n
T /2

T / 2
sin( m 0 t) cos(n 0 t)dt  0, for all m and n 20
 Continuous Discrete
Time Time

Periodic

Continuous Time Discrete Time

Fourier Fourier
Series (CTFS) Series (DTFS)

Discrete Time
Continuous
Fourier
Aperiodic

Time Fourier
Transform
Transform (CTFT)
(DTFT)

21
The Fourier Series

Joseph Fourier
1768 to 1830

22
 The Fourier Series
Fourier proposed in 1807

A Fourier Series is an accurate representation of a periodic

signal and consists of the sum of sinusoids at the
fundamental and harmonic frequencies.

A periodic waveform x(t) could be broken down into an

infinite series of simple sinusoids which, when added
together, would construct the exact form of the original
waveform.

23
 Periodic Functions
A function xt  is periodic
if it is defined for all real
and if there is some positive number,

T such that xt  T  . xt 

24
 f  f  

0 0
 

T
T

f 

0


25
 How Does FT Work Anyway?

 FS/FT uses complex exponentials (sinusoids) as

building blocks. e j t  cos  t   j sin t
 For each frequency of complex exponential, the sinusoid
at that frequency is compared to the signal.
 If the signal consists of that frequency, the correlation is
high  large FS coefficients.

 If the signal does not have any spectral component at a

frequency, the correlation at that frequency is low / zero,
 small / zero FT coefficient.
x1(t)  cos(25t)

x2 (t)  cos(2 25t)

x3 (t)  cos(2 50t)

x1(t) F X (

x( F X 2 ()

x3 (t) F X 3 ()
 Dirichlet conditions for existence of FS

1. x(t) Should be a single-value function except at

possibly a finite number of points.

2. The integral

for any t0.

3. x(t) should have finite number of discontinuities within
the period T.
4. x(t) should have a finite number of maxima and
minima within the period T.

29
 FS Synthesis equation

2nt  2nt
x(t)  a 0   an cos   bn sin
n1 T n1 T

DC Part Even Part Odd Part

T is a period of all the above signals

Let 0=2/T.
 
x(t)  ao   an cos(n0t)   bn sin( n0t)
30
n1 n1
FS Decomposition equation
 

x(t)  ao   an cos(n0t)   bn sin( n0t)

n1 n1

1 t0 T
a0 
T 
t0
x(t)dt

2 t 0T
an 
T  t0
x(t) cos n 0tdt n 1,2,

2 t 0 T
bn 
T  t0
x(t) sinn 0tdt n 1,2,

31
P1: Find the Fourier series of the following periodic
function.

f  
A

0


-A
        

f   A when 0   
 A when    2
f   2   f 
32
 1 t0 T
a0 
T 
t0
x(t)dt
1 2
  
2 0
a0  f d

1  
 
2
  

2 
 0
f  d   
f  d

1   2


2 
 0
Ad   
  A d

0 33
 2
f cos n d
1
an 
  0

1  2

  Acos n d    Acos n d
  0  
 2
1  sinn  1 sinn 
 A   A  0
 n 0   n 

34
 2
f sinn d
1
bn 
  0

1  2

   A sinn d    Asinn d
 0  
 2
1 cos n  1  cos n 
 A   A
  n 0   n 


A
cos n  cos 0  cos 2n  cos n 
n
35
 bn 
A
 cos n  cos 0  cos 2n  c
n

A
1 1  1  1
n
4A


bn 
A
cos n  cos 0  cos 2n  cos n 
n

A
1  1  1  1
n
21 Janu 2021 when n is even
ary0 36
 Therefore, the corresponding Fourier series is
 
x(t)  ao   an cos(n0t)   bn sin( n0 t)
n1 n1


x(t)  ( ) *  ( sin( nt))

4A 1
 n1 n

4 A 1 1 1
x(t)  sin   sin 3
 
How how many terms to consider?
37
When we consider 4 terms as shown in the
previous slide, the function looks like the
following.
1.5

0.5

f( ) 0

0.5

1.5

21 January 2021 38
When we consider 6 terms, the function looks
like the following.
1.5

0.5

f ( ) 0

0.5

1.5

39
When we consider 8 terms, the function looks
like the following.
1.5

0.5

f() 0

0.5

1.5


40
When we consider 12 terms, the function looks
like the following.
1.5

0.5

f( ) 0

0.5

1.5


41
 The red curve was drawn with 12 terms and
the blue curve was drawn with 4 terms.

1.5

0.5

0.5

1.5


21 January 2021 42
The red curve was drawn with 12 terms and
the blue curve was drawn with 4 terms.
1.5

0.5

0.5

1.5
0 2 4 6 8 10
 43
 The red curve was drawn with 20 terms and
the blue curve was drawn with 4 terms.
1.5

0.5

0.5

1.5
0 2 4 6 8 10
21 January 2021  44
21 January 2021 45
 Gibbs Phenomenon
1.5

0.5

0.5

1.5 0 2 4 6 8 10


The truncation of frequency content causes a time domain ringing artifact on

the signal, Any signal with a sudden discontinuity or step will show the Gibbs
phenomenon.
Gibbs' phenomenon occurs near a jump discontinuity in the signal. It says that
no matter how many terms you include in your Fourier series there will always
be an error in the form of an overshoot near the discontinuity. The overshoot
always be about 9% of the size of the jump.

21 January 2021 46
 MATLAB Code for Synthesis of FS

% Square wave ****************

clc;
clear all;
t=0:0.0001:0.1;
y=(4/pi)*(sin(t)+ 1/3*sin(3*t)+
1/5*sin(5*t)+1/7*sin(7*t)+1/9*sin(9*t));
plot(t,y1)

4 A 1 1 1 
s in   s in 3  s in 5  s in 7  
  3 5 7 
21 January 2021 47
% Square wave ****************
clc;
clear all;

)*
t=0:0.0001:0.1; 4A 1
n=1001; x(t)  (
sum=0;
for i=1:2:n
y1=(1/i*sin(i*t));
sum=y1+sum;
end
y1=(4/pi)*sum;
plot(t,y1);title('n=1001');xlabel('Time');ylabel('Amplitude');
Mag1=abs(fft(y,2048));
for j=1:1024
f(j)=j*10000/(2*1024); %fs/2
end
plot(f,Mag1(1:1024)); xlabel('Frequency in Hz'); ylabel('Magnitude');
plot(angle(y));%xlabel('Frequency in Hz') ;ylabel(‘phase’)
%***********************************************
4 A 1 1 1 
s in   s in 3  s in 5  s in 7  
 
21 January 2021
3 5 7  48
 Fourier series: example 1

21 January 2021 49
 Fourier series: example 2

21 January 2021 50
• Compare the plots of the partial sums with
the original function:

N 3
N 1

N 5 N 11
21 January 2021 51
 P2: Find TFS of Square Wave
f(t)
1

-6 -5 -4 -3 -2 -  2 3 4 5

2 


a0  1dt  1/ 2
0

2  1


an  cos ntdt  sin nt 0
0 n 1,2,
2 0 n

2  1  1  2 / n n 1,3,5,
bn 
2 
0
sin ntdt  
n
cos nt 0  
n
(cos n  1)  
0 n  2,4,6,

21 January 2021 52
 1 2 1 1 
f (t)   sin t  sin 3t  sin 5t  
2  3 5 

f(t)
1

-6 -5 -4 -3 -2 -  2 3 4 5

2 


a0  1dt  1/ 2
0

2  1 
an 
2 0
cos ntdt 
n
sin nt 0  0 n 1,2,

1  1  1 2 / n n 1,3,5,
bn 
2 0
sin ntdt  
n
cos nt 0  
n
(cos n  1) 
0 n  2,4,6,
21 January 2021 53
 f(t)
1

-6 -5 -4 -3 -2 -  2 3 4 5

2 

11dt0a
02

1.5

1 1 2 
,
21
,n
0s
n
i c
os ntdt nt
a 0.5
2

n 0 0n

b
/20
)1 
 -0.5
0
n  0n2 n0ncos(1

ncos
1 
1
ntsin 


ntdt n 1,3,5,
 n  2,4,6,
21 January 2021 54
 1 2  1 1 
f (t)    s i n t  sin 3t  sin 5 t  
2   3 5 
clc;
clear all;
t=0:0.0001:30;
f=2*pi;
x=sin(t);
%plot(t,x)
y= sin(t)+ 1/3*sin(3*t)+ 1/5*sin(5*t)+1/7*sin(7*t)+1/9*sin(9*t);
y=1/2+(2/pi)*y;
plot(t,y)

n=1000;
sum=0;
for i=1:2:n
z=1/i*sin(i*t);
sum=sum+z;
end
sum=1/2+(2/pi)*sum;
plot(t,sum);axis([0 30 0 1]);
21 January 2021 55
 Symmetry Properties
Even Functions

xt xt

21 January 2021 56
 Odd Functions

x   t    x t 

21 January 2021 57
 Even functions can solely be represented by cosine waves
because, cosine waves are even functions. A sum of even
functions is another even function.

5
10 0 10

21 January 2021 58
 Odd functions can solely be represented by sine waves
because, sine waves are odd functions. A sum of odd
functions is another odd function.

10 0 10

21 January 2021 59
 Properties – Symmetry

2 t 0T
an 
T  t0
x(t) cos n 0tdt n 1,2,

2 t 0 T
bn 
T t0
x(t) sinn 0tdt n 1,2,

21 January 2021 60
 Properties – Half Wave Symmetry
• If the function is half-wave symmetric, then only odd
harmonics exist

Half wave symmetry: f(t-T0/2) = -f(t)

-
A

-1 1 2

21 January 2021 61
 Properties – Discontinuities
• If the function has
– Discontinuities: the coefficients will be proportional to 1/n
– No discontinuities: the coefficients will be proportional to 1/n2

Which function has Which is closer to

discontinuities? a sinusoid?
-A

-1 1 2
A

-2 -1 0 1 2

21 January 2021 62
 Example
P3: Without any calculations, predict the general form of
the Fourier series of:

-
A

-1 1 2

DC? No, a0 = 0; Half wave Yes, only odd

symmetry? harmonics

Symmetry? Even, bn = 0; Discontinuities?No, falls of as

1/n2
Prediction an  1/n2
for n = 1, 3, 5, …;
21 January 2021 63
• Now perform the calculation
T0  2; 0  2  
2
2 t1 T0 4 T0 / 2
an 
T0 t1
 x(t) cos(n0t)dt 
T0 0
 x(t) cos(n0t)dt

zero for
1 n even
an  2  2At cos(n t)dt 
4A
1 cos(n )
0 n
2 2

8A
an  ...n 1,3,5...
n
2 2

21 January 2021 64
• tring_p3

21 January 2021 65
 Example
• Now compare your calculated answer with your
predicted form

DC?No, a0 = 0; Half wave Yes, only odd

symmetry? harmonics

SymmeEtrvy?en, bn = 0; Discontinuities?No, falls of as

1/n2

21 January 2021 66
 P 4. Find the Fourier series of the following
periodic function.
f(t)

3 T /4

0 t
T /4

- T /2 T /2 T 2T

f t   t when 
T
 t 
T
4 4
T T 3T
  t  w hen  t 
2 4 4

21 January 2021 67
 f t  T   f t

This is an odd function. Therefore, an  0

 2n
T

b n 
2
 f t  sin t dt
T 0  T 
T

 2n
2


4
 f t  sin t dt
T 0  T 

21 January 2021 68
 T

4 4
 2  n 
b n 
T  t s i n 
 T
t  d t

0
T

4
2
  t  T  s i n  2  n 

T   2   T
t  d t

T
4

Use integration by parts.

   n   
2
4  T
b   2 .   s in   
 2  n 
n
T   2 
2 T  n  
 s i n  
n 2
 2
 2 
bn  0 when n is even.
21 January 2021 69
 Therefore, the Fourier series is
Tring_p4.m

2T  2  1  6  1 10  
x(t)  in
2 s  t   2 sin t   2 sin t  
  T  3 T  5  T  

21 January 2021 70
P5: Find the coefficients for the Fourier
series of:

 t    t  0
x(t )  
 t 0  t  
x(t  2 )  x(t )

f (x)

Fundamental
range
x
0 2 4
Period = 2π

21 January 2021 71
 • Find a0
T
2 
1
 x(t)dt
1
a0 
T  x(t)dt  a0 
2
T 
2

f (x) is an even function so:


1
 x(t)dt

a0  1
2 
 a0 
  x(t)dt
0

1 1 t  2 
 a0   tdt  a0    
   2 0
 a0 
0 2

21 January 2021 72
 • Find an
T

2
 1  2nt 

2
an 
T
 x(t) c  an 
 
x(t) cos 
 2 
dt

Since both functions are even their product is even:


2
 x(t) cosntdt  t cosntdt
1
an   an 
 
 0

21 January 2021 73
• Find bn
T

2 2
 1  2nt 
bn 
T
 x(t)s  bn  
 
x(t)sin 
 2 
dt

Since sine is an odd function and x(t) is an even

function, the product of the functions is odd:

1
bn 
  x(t) sin ntdt

 bn  0

21 January 2021 74
 tring_p5.m

• So we can put the coefficients back into the

Fourier series formula:

 x(t)     2 1n 1cosnt 

 n

2 n1  n 2 

 x(t) 
  4 cost  0  4 cos3t  
2  9

21 January 2021 75
 The Complex Form of Fourier Series
 

f  a0  a cos n   b s inn

n n
n1 n1
Let us utilize the Euler formulae.

j  j
e e j
e  
cos   sin  
2

21 January 2021 76
 Complex Form of the Fourier Series

c0  ao
x(t)  
n  
cne jn 0t

1
cn  (an jb n)
T/ 2
2
1
cn   x(t)e dt
jn0t 1
cn  (an jb n)
T T/2 2
cn | cn | e j n , cn  c n* | cn | e  j n
If x(t) is real,
1 a2
c  c n  bn
2
* | cn || cn |
2 n  1,2,3,
n n
 b 
 n  tan 1   n  c0  a 0
21 January 2021
 an  77
 Perceval's Relationship
]

 
x(t)   cne j n  0t
x *(t)   c * n e  jn0t

n   n

21 January 2021 78
21 January 2021 79
 Complex Frequency Spectra
cn | cn | e j n , cn  c n* | c n | e  j n   bn 
n  tan 1 n  1,2
1  an 
| cn || cn |

c0  a 0
|cn| amplitude
spectrum

phase
n
spectrum


21 January 2021 80
 P6: Exponential FS
x(t)
A
t
 
T

 T T
2 2 2


2

A  /2 A 1
cn   e  jn0tdt  (2 j sin n0 /2)
T  / 2 T  jn 0
 /2 A 1
A 1  sin n 0 / 2
 e jn0t T 2n 0
1
T  jn 0
sin  n 
 / 2

A  T 
A 1  jn0 /2 1 jn0 / 2  
  e  e  T  n 
T  jn0
21 January 202 1
 jn0   
 T  81
 A/5

-120 -80 -40 0 40 80 120

-150 -100 -50 50 100 150

sin  n   ,
1 T  1, 

1
A  T 
cn  20 4 T 5
T  n  2
  0   8
 T  T

21 January 2021 82
 A/10

-120 -80 -40 0 40 80 120

-300 -200 -100 100 200 300

sin  n 
1  1
  1 , T , 
A  T  20 2 T 5
cn 
T  n 
  0  2  4
 T  T

21 January 2021 83
 Summary
 

x(t)  ao   an cos(n0t)   bn sin( n0 t)

n1 n1

1 t0 T
a0 
T  t0
x(t)dt

2 t 0T
an 
T  t0
x(t) cos n 0tdt n 1,2,

2 t 0 T
bn 
T t0
x(t) sinn 0tdt n 1,2,

21 January 2021 84
 Exponential Fourier Series

c0  ao
x(t)  
n  
cne jn 0t

1
cn  (an jb n)
T/ 2
2
1
cn   x(t)e dt
jn0t 1
cn  (an jb n)
T T/2 2
cn | cn | e j n , cn  c n* | cn | e  j n
If x(t) is real,
1 a2
c  c n  bn
2
* | cn || cn |
2 n  1,2,3,
n n
 b 
 n  tan 1   n  c0  a 0
21 January 2021
 an  85
 Perceval's Relationship

21 January 2021 86
 FT Shortcomings

 Complex exponentials stretch out to infinity in

time
– They analyze the signal globally, not locally
– Hence, FT can only tell what frequencies exist in the entire
signal, but cannot tell, at what time instances these
frequencies occur
– In order to obtain time localization of the spectral
components, the signal need to be analyzed locally, BUT
– HOW ?
 Sound spectrum

amplitude
Time f 3f 5f 7f
frequency
 Application of Fourier Transform for Music Signal
Example: Timbre
Timbre, also known as sound quality or tone color, is oddly
defined in terms of what it is not:
When two sounds are heard that match for pitch, loudness, and duration,
and a difference can still be heard between the two sounds, that
difference is called timbre.
There are two physical correlates of timbre:
Spectrum envelope
Amplitude envelope

89
 Six Synthesized Sounds Differing in Spectrum Envelope

Note the similarities in pitch (due to constant F0/harmonic spacing) and the
differences in timbre or sound quality.
Same melody, same spectrum envelope (if sustained), different
amplitude envelopes (i.e., different attack and decay characteristics).
Note differences in timbre or sound quality as the amplitude envelope
varies.

I
 Timbre and Spectrum Envelope

Timbre differences between one musical instrument and another are

partly related to differences in spectrum envelope.
 Outcome based education-
Summary
• Prepare global engineers who will have
to solve problems and shoulder
challenges which are not even known
today.
That means focus will have to be on
• Understanding fundamentals very well
• Learning new skills
• Competencies that enable individuals to
cope with the demands of the rapidly
changing workplace
 What Industry/OBE Expect?
Student should able to
•Write Project proposals
•Complete Project
•Analyse case studies
•Give case presentations
•Abilities to think
•Abilities to research
•Abilities to question
•Make decision based on findings
•Evaluate Projects
•Propose solution to problems
•Market their solutions
94
Blooms Taxonomy
 Outcome Based Education

• Miniprojects
• Create: Simulate different musical instruments: Guitar, Piano, Trumpet ……
• Analysis: Find Frequencies of male/Female/Children frequency
• Analysis/Justify: Analysis of voice Signals
• ( Why Females are preferred as anchoring?)
• Analysis of animal Voice
• Simulation of different Human and Animal Voice
• Extract Some features to distinguish voice (Animal/Human)
• Justify
• Compute
• Evaluate
• Show that
 Activities for advanced Learners

• Simulations of voice and writing of papers using FS/FT

• Create GUI
• Reading of Research Papers and Group discussion on that
• NPTEL/Swayam course
• Case studies presentation : Speech Recognition/
• Speaker Recognitions/Audio classification/Musical l Instrument classification
• Human speech Production Model
• Human Auditory System
 References

Dr. Vikram Gadre, IIT Mumbai

https://www.youtube.com/watch?v=NH6CFP1wGrw&list=PLjvLvSytzo
QPG5wSWGO-1Ae1BDStdCjvY&pbjreload=101

Principles of Signals & Systems: IIT Kanpur

https://nptel.ac.in/courses/108/104/108104100/

MIT Open Courseware : http://ocw.mit.edu/terms.

NPTEL : Signals & Systems in Hindi : IISER Bhopal

21 January 2021 98
Thanks

99