EEE4221 Digital Signal Processing

Lecture 3 : Discrete-Time Signals & Systems

Instructor: Dr D. Banda
Email: daliso.banda@unza.zm

March 2022
 References
Our main reference text books in this course are
[1] John G. Proakis, Dimitris Manolakis, “Digital Signal Processing:
Principles, Algorithms and Applications”, 4th Edition, (2006), Prentice
Hall, ISBN-10: 0131873741, ISBN-13: 978-0131873742.
[2] Alan V. Oppenheim, Ronald W. Schafer, John R. Buck, Discrete-Time
Signal Processing, 3rd Edition, (2010), Prentice-Hall, ISBN: 0137549202 /
0-13-754920-2.
[3] DIMITRIS G. MANOLAKIS, VINAY K. INGLE, Applied Digital Signal
Processing, 2nd Edition, (2011) Cambridge University Press, ISBN 978-0-
521-11002-0.

However, feel free to use some additional text which you might find relevant
to our course.
Based on notes produced by Jerry MUWAMBA , DEPT. of EEE, School of
Engineering, UNZA
2
 Discrete-Time Signals: Sequences
 Discrete-time signals can be represented by sequence of numbers
 The nth number in the sequence is represented with x[n]
 Often times sequences are obtained by periodic sampling of
continuous-time signals
 In this case x[n] is value of the analog signal at xc(nT)
 Where T is the sampling period (fs=1/T)

10

-10
0 20 40 60 80 100 t (ms)
10
T
0

-10
0 10 20 30 40 50 n (samples)

3
 Discrete-Time Signals: Sequences
 Discrete-time signals can also be can also represent images
 They can be represented by sequence of numbers usually referred
to as pixel values.

4
 Discrete-Time Signals: Representations
• Apart from Sequence representation, there are several ways to represent a
discrete-time signal.
• The more widely used representations are illustrated below:

5
 Basic Sequences

 Delayed or Shifted version of a sequence x[n]

y[n]  x[n  no ]

 Arbitrary sequence expressed as

sum of scaled delayed impulses

x[n]   x  k   n  k 
k 

6
 Basic Sequences

 Unit sample (impulse) sequence 1.5

Graphical representation

0, n  0
 [n]  
1

1, n  0 0.5

0
-10 -5 0 5 10
 Unit step sequence 1.5

0 , n  0 1
u[ n]  
1, n  0 0.5

0
-10 -5 0 5 10
1
 Exponential sequences
0.5
x[n]  A n

0
-10 -5 0 5 10

7
Basic Operations

8
 Basic Operations
 Slightly more complicated operation involves transformations of the independent
variable n.
 Two important time-based transformations are:
1) Time-reversal or folding:
 An operation defined by y[n] = x[−n], reflects the sequence x[n] about the origin n = 0.
 This time-reversal operation, which obviously cannot be done in real-time.
 If x[−n] = x[n] the sequence is called even or symmetric
 If x[−n] = −x[n] it is called odd or antisymmetric.

9
 Basic Operations
 Slightly more complicated operation involves transformations of the independent
variable n.
 Two important time-based transformations are:
2) Time-shifting :
 Time-shifting is defined by the formula y[n] = x[n − n0].
 For n = n0 we have, y[n0] = x[0]; thus, the sequence x[n] is shifted by n0 samples so that the sample
x[0] is moved to n = n0.
 If n0 > 0, the sequence x[n] is shifted to the right; because the sequence “appears later,” the shift
corresponds to a time-delay.
 If n0 < 0, the sequence is shifted to the left; because the sequence “appears earlier,” the shift amounts
to a time-advance.

10
 Example 3.1: Combining Basic Sequences
 We most often combine basic sequences to form simple representations
of other sequences.
 If we want an exponential sequence that is zero for n < 0, we can write
a somewhat cumbersome expression as

 A n , n  0
x  n  
0 , n0

 A much simpler expression is x  n   A nu  n 

11
 Sinusoidal Sequences
 Are also very important. In general the sequence is of the form

x  n   A cos o n    ,  n

Where:
A (amplitude) and φ (phase) are real constants.
ω0 is the frequency of the sinusoid and has units of radians per sampling interval.

12
 Sinusoidal Sequences

 For an exponential sequence with complex    e j and A  A e j o

x  n   A n  A e j  e jo n  A  e
n n j o n  

x  n   A  cos o n     j A  sin o n   

n n

 x[n] is a sum of exponentially weighted sinusoids

 Different from continuous-time, in that discrete-time sinusoids
 Have ambiguity of 2k in frequency

x  n  A cos o  2 k  n     cos o n   

 Are not necessary periodic with 2/o

13
 Sinusoidal Sequences
 Thus, when discussing complex exponentials signals of the form
x  n   Ae j n or real sinusoidal signals of the form x  n   cos o n   
0

, we need only consider frequencies in an interval of length 2 ,

such as   0   or 0  0  2 .

14
 Sinusoidal
 Thus, in an interval of length , Sequences
such as
.
 In the discrete-time case, a periodic sequence is a kind for which

x  n  x  n  N  ,  n

where the period N is necessarily an integer.

 The discrite-time sinusoidal is periodic if
2 k
cos o n     cos o n  o N    iff N is an integer
o
where k is an integer.

15
 Example 3.2: Periodic and Aperiodic Discrete-
Time Sinusoids
 Consider the signal x1  n   cos  n 4  . This signal has a period of
N  8 . To show this, note that
x  n  8  cos   n  8  4   cos  n 4  2   cos  n 4   x  n 

 This satisfies the definition of a discrete-time periodic signal.

16
 Example 3.2: Periodic and Aperiodic Discrete-
Time Sinusoids
 Increasing the frequency of a discrete-time sinusoid does not
necessarily decrease the period of signal
 Consider the discrete sinusoid x2  n   cos  3 n 8  , with a higher freq
than x1  n  . However, x2  n  is not periodic with period 8, since
x2  n  8  cos  3  n  8  8   cos  3 n 8  3    x2  n 
 Using an argument analogous to the one for x1  n  , it can be shown
that x2  n  has a period of N  16 .
 Notice the increase in period with increase in freq since n is integer.

17
 Discrete-Time Systems
 Discrete-time system is a mathematical operation (transformation) that
maps a given input sequence x[n] into an output sequence y[n]

 Example 3.3 Discrete-Time Systems 1 M2

y[n]   x n  k 
 Moving (Running) Average M1  M 2  1 k  M1
System computes the nth sample of the output sequence as the
average of  M 1  M 2  1 samples of the input sequence around the
nth sample.
 Maximum y[n]  max  x[n], x[n  1], x[n  2]

 Ideal Delay System y[n]  x[n  nd ]

18
 Memoryless Systems
 Memoryless System
 A system is memoryless if the output y[n] at every value of n depends only
on the input x[n] at the same value of n

 Example 3.4 Memoryless Systems

y[n]   x[n]
2
 Square

 Sign/Signum y[ n]  sgn  x[n]

 Counter Example
 Ideal Delay System y[n]  x[n  nd ]

19
 Linear Systems
 Linear System: A system is linear if and only if the two properties hold,
T {x1[n]  x2 [n]}  T  x1[n]  T  x2 [n] (additivity property )

T ax[n]  aT  x[n]  ay  n  ( hom ogeneity or scaling property )

 In condensed form, we thus have the principle of superposition

T {ax1[n]  bx2 [n]}  aT  x1[ n]  bT  x2 [ n]

 Example 3.5
 Ideal Delay System y[n]  T  x  n  x[n  nd ]
T {ax1[n]  bx2 [n]}  ax1[n  nd ]  bx2 [n  nd ]
aT {x1[n]}  bT  x2 [n]  ax1[n  nd ]  bx2 [n  nd ]
 Vividly, the principle of superposition holds, hence the system is
Linear.

20
Linear Systems: Basic Building Blocks

21
 Time-Invariant Systems
 Time-invariant (shift-invariant) Systems
 A time shift at the input causes corresponding time-shift at output

y[n]  T {x[n]}  y[ n  no ]  T  x[ n  no ]

 Example 3.6
 Square
y1  n    x[n  no ]
2
Delay the input the output is
y[n]   x[n]
2

y  n - no    x[n - no ]
2
Delay the output gives
 Counter Example
 Compressor/Downsampler System
Delay the input the output is y1  n   x[ Mn  no ]
y[n]  x[ Mn]
Delay the output gives y  n - no   x  M  n  no  

22
 Causal System
 Causality
 A system is causal if it’s output is a function of only the current and
previous samples

 Example 3.7
 Backward Difference
y[n]  x[n]  x[n  1]

 Counter Example
 Forward Difference
y[n]  x[n  1]  x[n]

23
 Stable System
 Stability in the sense of bounded-input bounded-output (BIBO)
 A system is stable if and only if every bounded input sequence produces a
bounded output sequence.
x[n]  Bx    y[n]  By   ,  n

 Example 3.8
y[n]   x[n]
2
 Square
if input is bounded by x[ n]  Bx  
output is bounded by y[ n]  x[ n]  Bx2  
2

 Counter Example
 Log y[n]  log10  x[n] 

even if input is bounded by x[n]  Bx  

 
output not bounded for x  n   0  y  n   log10 x  n   

24
 End of Lecture 3

Thank you for your attention!

25