EE331 Signals & Systems Chapter # 3

EE331
Signals & Systems
Chapter # 3
Fourier Series Representation
of Periodic Signals

15-16-18.10.2024 1
EE331 Signals & Systems Chapter # 3 Introduction
In Chapter 2, we represented signals as linear combinations of shifted impulses.

In this Chapter, we will represent signals as linear combinations of another set of

basic signals, named complex exponentials.

The resulting representations are known as Fourier Series (Chapter 3) and Fourier
Transform (Chapter 4 and 5).

Complex exponential signals are the signals of the form:

est in CT
zn in DT where s and z are complex numbers.

15-16-18.10.2024 2
 3.2 The Response of LTI Systems to Complex
EE331 Signals & Systems Chapter # 3 Exponentials

Please read Section 3.2!

The response of a LTI system to a complex exponential input is the same complex
exponential with only a change in amplitude.

CT : e st  H ( s )e st 

DT : z n
H ( z) z n

eigenfunction of the system 
H(s) and H(z) are the system’s eigenvalue.

H(s) or H(z) are the complex amplitude factors and are generally a function
of the complex variable s or z.

15-16-18.10.2024 3
 3.2 The Response of LTI Systems to Complex
EE331 Signals & Systems Chapter # 3 Exponentials

To show complex exponentials are eigenfunctions of LTI systems:

If x(t) = est
   
y (t )  

h( ) x(t   )d   h( )e s (t  ) d   h( )e st e  s d  e st
 


h( )e  s d  e st H (s )  H (s )e st

H (s)

• H(s): The complex constant depends on s and which is related to the system impulse
response

• The constant H(s) for a specific value of s is then the eigenvalue associated with the
eigenfunction est.

Similarly, it can be shown that zn are eigenfunctions of DT LTI systems.

15-16-18.10.2024 4
 3.2 The Response of LTI Systems to Complex
EE331 Signals & Systems Chapter # 3 Exponentials

For both CT and DT, if the input to a LTI system is represented as a linear
combination of complex exponentials, then the output can also be represented as a
linear combination of the same complex exponentials.

Example:
x(t )  a1es1t  a2es2t  a3es3t
From the eigenfunction property the response for each part:
a1e s1t  a1H ( s1 )e s1t
a2e s2t  a2 H ( s2 )e s2t 
From the superposition property
 y (t )  a1H (s1 )e s1t  a2 H (s2 )e s2t  a3 H (s3 )e s3t
a3e s3t  a3 H ( s3 )e s3t
In general,
x(t )   ak esk t   y(t )   ak H (sk )e sk t
k k
15-16-18.10.2024 5
 3.2 The Response of LTI Systems to Complex
EE331 Signals & Systems Chapter # 3 Exponentials

Note: In general, s and z are arbitrarily complex numbers. Fourier analysis restricts
our attention to particular forms for these variables.

In CT: s = jω (pure imaginary), Complex exponential: ejωt

In DT: z = ejω (unit magnitude), Complex exponential: ejωn

Example 3.1

a) y(t) = x(t-3)

If x(t) = ej2t, then y(t) = ej2(t-3) = e j 6  e j 2t

eigenvalue H ( j 2) eigenfunction

15-16-18.10.2024 6
 3.2 The Response of LTI Systems to Complex
EE331 Signals & Systems Chapter # 3 Exponentials

To confirm:

h(t )   (t  3)



 s
H (s)   (  3) e d  e 3 s
so that H ( j 2)  e  j6



b) x(t )  cos(4t )  cos(7t )  y (t )  cos(4(t  3))  cos(7(t  3))

Using Euler’s relation:

1 j 4t 1  j 4t 1 j 7 t 1  j 7 t
x(t )  e  e  e  e 
2 2 2 2
1  j12 1 j12  j 4t 1  j 21 j 7 t 1 j 21  j 7 t
y (t )  e e  e
j 4t
e  e e  e e
2 H ( j 4) e j12 2 H ( j ( 4)) 2 H ( j 7) 2 H (  j 7)
eigenvalue

15-16-18.10.2024 7
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals
3.3 Fourier Series Representation of CT Periodic Signals

3.3.1 Linear Combinations of Harmonically Related Complex Exponentials

A signal is periodic if, for some positive value of T:

x(t) = x(t + T) for all t.

Fundamental period: The minimum positive, nonzero value of T.

Fundamental frequency: ω0 = 2π / T

x(t )  cos 0t : The sinusoidal 

 Periodic with fund freq 0 and fund period T  2 / 0
x(t )  e j0t : Periodic complex exponential 

15-16-18.10.2024 8
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

The set of harmonically related complex exponentials associated with x(t )  e j0t :
k (t )  e jk t  e jk (2 / T )t , k  0, 1, 2,...
0

Each of these signals has a fundamental frequency that is multiple of 0, and therefore,
each is periodic with period T.
Thus, a linear combination of harmonically related complex exponentials of the form:
 
x(t )  
k 
ak e jk0t
  ak e jk (2 / T )t
k 
(3.25)
is also periodic with period T.
The representation of a periodic signal in the form of Eq. 3.25 is referred to as the
Fourier Series representation.

15-16-18.10.2024 9
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

In Eq. 3.25;

• The term for k = 0 is a constant

• The terms for k = ±1 both have fundamental frequency equal to 0 are referred
to as the fundamental components or the first harmonic components.

• The terms for k = ±2 are periodic with half the period of the fundamental
components and are referred to as the second harmonic components.

• More generally, the components for k = ±N are referred to as the Nth harmonic
components.

15-16-18.10.2024 10
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

Example 3.2

x(t) is periodic with fundamental frequency 2π.

3

 jk 2  t 1 1 1
x(t )  a k e with a0  1, a1  a1  , a2  a2  , a3  a3 
k 3 4 2 3
Rewriting,

x(t )  1  e
4

1 j 2 t
e 
 j 2 t
 e
2

1 j 4 t
e  j 4 t
 
1 j 6 t
 e
3
 e  j 6 t  or

1 2
x(t )  1  cos 2 t  cos 4 t  cos 6 t  Alternative form for the Fourier
2 3
Series of real periodic signals.

15-16-18.10.2024 11
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

15-16-18.10.2024 12
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

15-16-18.10.2024 13
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

The other alternative forms of the Fourier Series:


x(t )  a0    ak e jk0t  a k e  jk0t  or
k 1



x(t )  a0   2 Re ak e jk0t  or
k 1


or (using ak  Ak e j k
): x(t )  a0  2 Ak  cos( k0t   k )
k 1


or (using ak  Bk  jCk ) : x(t )  a0  2  Bk  cos k0t  Ck  sin k0t 
k 1
15-16-18.10.2024 14
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

3.3.2 Determination of the Fourier Series Representation of a CT Periodic

Signal
 

Synthesis Equation x(t )  

k 
ak e jk0t
  ak e j k ( 2  / T ) t
k 

1 1
 
 jk0t  jk (2 / T ) t
Analysis Equation ak  x (t ) e dt  x (t ) e dt
TT TT

The set of coefficients {ak} are often called the Fourier Series coefficients or the
spectral coefficients of x(t).

a0 : DC or constant component of x(t).

1
a0   x(t )dt (Average value of x(t) over one period.)
15-16-18.10.2024 TT 15
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

Example 3.3: x(t )  sin 0t which has fundamental frequency 0. To determine
Fourier coefficients, we may use analysis equation. But instead let’s try inspection
method:
1 j0t 1  j0t 1 1
x(t )  sin 0t  e  e  a1  , a1  , ak  0 for k  1
2j 2j 2j 2j
Example 3.4:

x(t )  1  sin 0t  2 cos 0t  cos(20t  ) which has fundamental frequency 0.
4
By inspection:
 
1 1  j ( 20t  )  j ( 20t  ) 
x(t )  1  e j0t  e  j0t   e j0t  e  j0t   e 4
e 4

2j 2 
15-16-18.10.2024 16
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

 1  j0t  1   j0t  1 j 4  j 20t  1  j 4   j 20t

x(t )  1  1   e  1   e   e e   e e 
 2j  2j 2  2 
1 1 1 1 1 j 4 2 1  j 4 2
a0  1, a1  1   1  j , a1  1   1  j , a2  e  (1  j ), a2  e  (1  j )
2j 2 2j 2 2 4 2 4

ak  0, | k | 2

15-16-18.10.2024 17
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

Example 3.5: 1, | t | T1

x(t )  
0, T1 | t | T / 2

periodic with fundamental period T and fundamental frequency 0 = 2π / T.

Using Analysis Equation:

Because of symmetry, choose –T/2 ≤ t < T/2 for integral interval.

15-16-18.10.2024 18
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

For k = 0;
T1
1 2T1
a0   1  dt 
T T1 T
For k ≠ 0;
 e jk0T1  e  jk0T1 
T1
1 1 2 2

 jk0t  jk0t T1
ak  1  e dt   e    sin(k0T1 )
T T1
jk0T T1 k0T  2j  k0T

2 sin(k0T1 ) sin(k0T1 )
ak   , k0
k0T k

Please look at Figure 3.7. on the next page.

Fourier Series can be used to represent an extremely large class of periodic signals.
15-16-18.10.2024 19
EE331 Signals & Systems Chapter # 3 3.3 Fourier Series Representation of CT Periodic Signals

15-16-18.10.2024 20
EE331 Signals & Systems Chapter # 3 3.4. Convergence of the Fourier Series
3.4. Convergence of the Fourier Series

Dirichlet conditions guarantees that x(t) equals its Fourier Series representation,

except at isolated values of t for which x(t) is discontinuous. At these values, the
  
infinite series of eq.   ak e jk0t  converges to the average of the values on either
 k  
side of the discontinuity.

15-16-18.10.2024 21
EE331 Signals & Systems Chapter # 3 3.4. Convergence of the Fourier Series
The Dirichlet Conditions:

Condition 1: Over any period, x(t) must be absolutely integrable.

 x(t ) dt  
T
 this guarantees that each coefficient ak will be finite.

1
Example: x(t )  , 0  t  1 violates this condition.
t

15-16-18.10.2024 22
EE331 Signals & Systems Chapter # 3 3.4. Convergence of the Fourier Series
Condition 2: In any finite interval of time, x(t) is bounded variation; that is there are
no more than a finite number of maxima and minima during any single period of the
signal.

 2 
Example: x (t )  sin   , 0  t  1  meets Condition 1 but violates Condition 2.
 t 

15-16-18.10.2024 23
EE331 Signals & Systems Chapter # 3 3.4. Convergence of the Fourier Series
Condition 3: In any finite interval of time, there are only a finite number of
discontinuities. Furthermore, each of these discontinuities is finite.

Example:

15-16-18.10.2024 24
EE331 Signals & Systems Chapter # 3 3.5. Properties of CTFS
3.5. Properties of CTFS
S
x(t )  ak
3.5.1. Linearity
S
x(t )  ak  S
S   z (t )  A  x (t )  B  y (t )  ck  A  ak  B  bk
y (t )  bk 

3.5.2. Time Shifting

S
x(t  t0 )  bk  e jk0t0 ak (When the signal is shifted in time the magnitudes of
its FS coefficients remain unaltered. That is |ak|=|bk|)

3.5.3. Time Reversal

S
x(t )  ak
15-16-18.10.2024 25
EE331 Signals & Systems Chapter # 3 3.5. Properties of CTFS
3.5.4. Time Scaling 
x( t )  
k 
ak e jk (0 )t 

FS coefficients have not changed. FS representation has changed because of the

change in fundamental frequency.

3.5.5. Multiplication

S
x(t )  y (t )  hk  ak  bk   ab
l 
l k l (As discrete time convolution)

3.5.6. Conjugation & Conjugate Symmetry

S
x* (t )  a* k

15-16-18.10.2024 26
EE331 Signals & Systems Chapter # 3 3.5. Properties of CTFS
3.5.7. Parseval’s relation for CT periodic signals


1

2

S

2
x(t ) dt ak
TT k 

Average power (Energy per unit time)in one period of the x ( t ) Average power

3.5.9. Examples (3.6., 3.7., 3.8., and 3.9.)

Please do them!

15-16-18.10.2024 27
EE331 Signals & Systems Chapter # 3 3.5. Properties of CTFS
3.5.8. Summary of Properties of the CTFS

15-16-18.10.2024 28
EE331 Signals & Systems Chapter # 3 3.5. Properties of CTFS

15-16-18.10.2024 29
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

3.6. FS Representation of DT Periodic Signals

The main difference between FS representation of DT periodic signals and FS

representation of CT periodic signals:

FS representation of DT periodic signals is finite series.

FS representation of CT periodic signals is infinite series.

3.6.1. Linear Combinations …

Read it! (Similar to 3.3.1)

15-16-18.10.2024 30
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

3.6.2. Determination of the FS Representation of Periodic Signal

Synthesis Equation x[n]  

k  N 
ak e jk0 n  
k  N 
ak e jk ( 2  / N ) n

1 1
Analysis Equation ak 
N

n  N 
x[n]e  jk0 n

N

n  N 
x[n]e  jk (2 / N ) n

The set of coefficients {ak} are often called the spectral coefficients of x[n].

15-16-18.10.2024 31
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

Example 3.10: x[n] = sin 0n

Note: x[n] is periodic only if 2π / 0 is an integer (i) or a ratio of integers (ii).

i) For the case 2π / 0 is an integer N that is 0 = 2π / N: x[n] is periodic with

fundamental period N.
1 j ( 2 / N ) n 1  j ( 2 / N ) n 1 1
x[n]  e  e  a1  , a1 
2j 2j 2j 2 j and the
remaining coefficients over the interval of summation are zero. These
1 1
coefficients repeat with period N. Thus N 1 a  , a N 1 
2j 2j

15-16-18.10.2024 32
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

For N = 5 (That is Fourier coefficients for x[n] = sin (2π/5)n ):

Note that only one period is utilized in the synthesis equation.

ii) For the case 2π / 0 is a ratio of integers, that is 0 = 2πM / N: Assuming M

and N do not have any common factors, x[n] has a fundamental period of N.

1 jM (2 / N ) n 1  jM (2 / N ) n 1 1
x[n]  e  e  aM  , a M  and the
2j 2j 2j 2j
remaining coefficients over one period of length N are zero.
15-16-18.10.2024 33
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

For M = 3 and N = 5 (That is Fourier coefficients for x[n] = sin 3(2π/5)n ):

Note that only one period (only two nonzero coefficients) is utilized in the
synthesis equation.

15-16-18.10.2024 34
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals
 2   2   4 
Example 3.11: x[n]  1  sin   n  3cos   n  cos  n 
 N   N   N 2
This signal is periodic with period N.

Thus the FS coefficients:

3 1 3 1 3 1 3 1 1 1
a0  1, a1     j, a1     j , a2  j , a2   j
2 2j 2 2 2 2j 2 2 2 2
With ak = 0 for other values of k in the interval of summation in the synthesis
equation.

Note that a-k = ak* for all values of k. In fact, this equality holds whenever x[n] is
real.

15-16-18.10.2024 Look at Figure 3.15 for Re{ ak }, Im{ ak }, | ak |, ∢ak 35

EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

15-16-18.10.2024 36
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

15-16-18.10.2024 37
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

Example 3.12:

N1
1 1
ak 
N

n  N 
x[n]e jk0n 
N

n  N1
1 e jk (2 / N ) n

N 1
 N,  1

Remember,   n
 1   N
  1  : any complex number
n 0 
 1

Letting m = n + N1

15-16-18.10.2024 38
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals
2 N1
1 1 jk (2 / N )N1 2 N1  jk (2 / N )m
ak 
N
 e
m 0
 jk (2 / N )( m  N1 )
 e
N

m 0
e

1 jk (2 / N ) N1  1  e  jk 2 (2 N1 1) / N 
ak  e   jk (2 / N ) 
N  1  e 
 jk (2 / 2 N ) jk 2 ( N1 1/ 2) / N  jk 2 ( N1 1/ 2) / N
1 e 
 e  e 
ak 
N e jk (2 / 2 N ) e jk (2 / 2 N )  e  jk (2 / 2 N ) 

 1 sin[2 k ( N1  1/ 2) / N ]
 N , k  0,  N , 2 N ,...
sin( k / N )
ak  
 2 N1  1
, k  0,  N , 2 N ,...
 N
Look for Fig. 3.17 for the coefficients ak for 2N1+1=5 for N=10, 20, and 40.
15-16-18.10.2024 39
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

15-16-18.10.2024 40
EE331 Signals & Systems Chapter # 3 3.6. FS Representation of DT Periodic Signals

Note: There are no convergence

issues with DTFS in general. The
reason is that any DT periodic
sequence x[n] is completely
specified by a finite number N of
parameters, the values of the
sequence over one period (Look Fig.
3.18). (Remember that this issue is
different in CTFS!)

15-16-18.10.2024 41
EE331 Signals & Systems Chapter # 3 3.7. Properties of Discrete-Time Fourier Series

15-16-18.10.2024 42
EE331 Signals & Systems Chapter # 3 3.7. Properties of Discrete-Time Fourier Series

15-16-18.10.2024 43
EE331 Signals & Systems Chapter # 3 3.7. Properties of Discrete-Time Fourier Series
Relatively different properties:
S
x[n] 

Please do examples 3.13., 3.14., 3.15!

ak
3.7.1 Multiplication:
S
x[n]  ak 
S   x[ n ]  y[ n ] S
 d k   al  bk l
y[n]  bk  l  N 

(Periodic convolution. See Problem 3.57)

3.7.2 First Difference

S
x[n]  x[n  1]  
1  e jk (2 / N )  ak 
3.7.3 Parseval’s Relation
1
 
2

2
x[n] ak
N n  N  k  N 
15-16-18.10.2024 Average power in one period Average power 44
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
3.8. Fourier Series and LTI Systems

Remember from Section 3.2

Similarly,

15-16-18.10.2024 45
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
When s or z are general complex numbers, H(s) and H(z) are called the system
function.

In the following chapters, we focus on the specific cases in which

Re{s} = 0 and |z| = 1

Then; s = jω ⟹ H(jω) is called frequency response of the system.


H ( j )  

h(t )e j dt

z = ej ⟹ H(z) = H(ej) is called frequency response of the system.


H ( z )  H (e j )  
n 
h[ n]e  j n

15-16-18.10.2024 46
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
As a result of superposition property, we can express the response of a LTI system to
a linear combination of complex exponentials:

CT case: x(t) is a periodic signal.

 
x(t )  
k 
ak  e jk0t y (t )  
k 
ak  H ( jk0 )  e jk0t with sk = jk0

y(t) is also periodic with the same fundamental frequency as x(t).

If ak is the set of FS coefficients for x(t), then {ak H(jk0)} is the set of FS
coefficients for y(t).
15-16-18.10.2024 47
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
Example 3.16
3

 k jk0t 1 1 1
x (t )  a  e , 0  2 a0  1, a1  a1  , a2  a2  , a3  a3 
k 3 4 2 3
h(t) = e-t · u(t) ⟹ y(t) = ?
 
1   j 1
Answer: H ( j )   e e
  j
d  e e 
0
1  j 0
1  j
0 = 2π then
3
1 1  1 1 
y (t )   k
b e jk 2 t
with bk = ak H(jk2π) ⟹ b0  1, b1  
4  1  j 2
 , b1  
4  1  j 2
,
k 3  
1 1  1 1  1 1  1 1 
b2    , b2    , b    , b3   
2  1  j 4  2  1  j 4  3  1  j 6  3  1  j 6 
3

The result can also be obtained in different forms. Please see the book!
15-16-18.10.2024 48
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
DT case: x[n] is a periodic signal.

x[n]  
k  N 
ak  e jk (2 / N ) n y[n]  
k  N 
ak  H (e j 2 k / N )  e jk (2 / N ) n with zk = ejk(2π/N)

y[n] is also periodic with the same period as x[n].

Thus the kth FS coefficient of y[n] is the product of the kth FS coefficient of the input
and the value of the frequency response of the LTI system, H(ej2πk/N), at the
corresponding frequency.
15-16-18.10.2024 49
EE331 Signals & Systems Chapter # 3 3.8. Fourier Series and LTI Systems
Example 3.17
 2 n  1 j (2 / N ) n 1  j (2 / N ) n
h[n] = αn u[n], -1 < α < 1 and x[ n ]  cos   e  e ⟹ y[n] = ?
 N  2 2

 
   e  1
H (e )    e
n
j n  j n  j
Answer: 
n 0 n 0 1   e  j
1 j (2 / N ) j (2 / N ) n 1
y[n]  H (e )e  H (e  j (2 / N ) )e  j (2 / N ) n
2 2
1 1  j (2 / N ) n 1  1   j (2 / N ) n
y[n]    e   e
2  1   e  j (2 / N )  2  1   e j (2 / N )

3.9. Filtering: Please read it!
3.10., 3.11. Read them!
15-16-18.10.2024 50