Discrete-Time Signals & Systems

1
 Review

Signals

Continuous-time Discrete-time

Continuous-value Discrete-value
Continuous-value

Analog Discrete Digital

2 Department of CSE, CUET

 Contents
 Discrete-Time Signals
 Representations
 Basic signals
 Classifications
 Manipulations
 Discrete-Time Systems
 Classification of Discrete-Time Systems
 Input-Output Description of Discrete-Time Systems
 Block Diagram Description of Discrete-Time Systems
 Interconnection of Discrete-Time Systems
 Analysis of Discrete-time linear time-invariant systems
 Techniques for Analysis of Linear Systems
 Resolution of Discrete-Time signal into impulse
 Response to Arbitrary Inputs: The convolution Sum
 Properties of Convolution of LTI System
 Causality and Stability of LTI System
 Finite vs. Infinite-Duration LTI System Response
 Discrete –Time Systems Described by Difference system
 Implementation of Discrete-Time Systems
 Correlation of Discrete-Time Signals

3 Department of CSE, CUET

 Discrete-Time Signals
 Signals are represented as sequences of numbers, called
samples
 Sample value of a typical signal or sequence denoted as,
𝑥 = *𝑥,𝑛-+ with −∞ < 𝑛 < ∞
 x[n] is defined only for integer values of n and undefined for non-
integer values of n
 Representation of discrete-time signals:
 Functional representation

 Tabular representation
n … -2 -1 0 1 2 3 …

x(n) … 0 0 0 1 4 1 …
 Sequence representation
x(n)={… , 0, 0, 0, 1, 4, 1, …}

4 Department of CSE, CUET

 Discrete-Time Signals

 Graphical representation

5 Department of CSE, CUET

 Basic Discrete-Time Signals

 Unit sample sequence

 Unit step sequence

 Unit ramp signal

6 Department of CSE, CUET

 Basic Discrete-Time Signals

 Real exponential signal

𝑥 𝑛 = 𝑎𝑛 for all n
Here, a is real

7 Department of CSE, CUET

 Basic Discrete-Time Signals

 Complex exponential signal

𝑥 𝑛 = 𝑟 𝑛 𝑒 𝑗𝜃𝑛 = 𝑟 𝑛 (cos 𝜃𝑛 + 𝑗 sin 𝜃𝑛)
Graphically represented by

𝑥𝑅 (𝑛) ≡ 𝑟 𝑛 cos 𝜃𝑛

𝑥𝐼 (𝑛) ≡ 𝑟 𝑛 sin 𝜃𝑛

𝜋
Here, r=0.9 and 𝜃 = 10

8 Department of CSE, CUET

 Basic Discrete-Time Signals
Graphically represented by

Amplitude function,
𝑥(𝑛) = 𝐴(𝑛) ≡ 𝑟 𝑛

Phase function,
∠𝑥 𝑛 = ∅ 𝑛 ≡ 𝜃𝑛

9 Department of CSE, CUET

 Classification of Discrete-Time Signals

 Energy signals and power signals

 The total energy of a signal x(n) is defined by

o Energy of a signal can be finite or infinite

o If E is finite (i.e. 0 < 𝐸 < ∞) 𝑥(𝑛)𝑖𝑠 𝑐𝑎𝑙𝑙𝑒𝑑 𝑎𝑠 an

energy signal
 The average power of a discrete-time signal x[n] is
defined by

10 Department of CSE, CUET

 Classification of Discrete-Time Signals
 Energy signals and power signals
 Define the signal energy of x(n) over the finite interval

- N ≤ n ≤ N as

 The signal energy can then be expressed as

 The average power of x(n) becomes

o If E is finite, P = 0. On the other hand, if E is infinite, the average

power P may be either finite or infinite
o If P is finite (and nonzero), the signal is called a power signal

11 Department of CSE, CUET

 Classification of Discrete-Time Signals

 Periodic signals and aperiodic signals

 A signal is periodic with period N (N > 0) if and only

if

The smallest value of N for which the above condition

holds is called the fundamental period

 A signal not satisfying the periodicity condition is

called nonperiodic or aperiodic
12 Department of CSE, CUET
 Classification of Discrete-Time Signals

 Symmetric (even) and antisymmetric (odd) signals

 A real-valued signal x(n) is called symmetric (even) if
x(-n)=x(n)

 A signal x(n) is called antisymmetric (odd) if x(-n)=-x(n)

13 Department of CSE, CUET

 Manipulation of Discrete-Time Signals

 Transformation of independent variable (time)

 Time shifting: A signal x[n] may be shifted in time by
replacing the independent variable n by n – k

• Signal x(n)

• Signal x(n-3) obtained delaying x(n) by 3 units in time

• Signal x(n+2) obtained advancing x(n) by 2 units in time

14 Department of CSE, CUET

 Manipulation of Discrete-Time Signals

 Transformation of independent variable (time)

 Folding/Reflection: A signal x[n] may be folded in
time by replacing the independent variable n by –n

15 Department of CSE, CUET

 Manipulation of Discrete-Time Signals

 Transformation of independent variable (time)

 Folding/Reflection
 The operations of folding and time delaying (or
advancing) a signal are NOT commutative
 Denote the time-delay operation by TD and the folding
operation by FD
An operation ʘ is
commutative if and only if
x ʘ y=y ʘ x for each x
and y.

16 Department of CSE, CUET

 Manipulation of Discrete-Time Signals
 Transformation of independent variable (time)
 Time Scaling or down-sampling: A signal x[n] may
be scaled in time by replacing n by µn

• Here, 𝑥 𝑛 = 𝑥𝑎 (𝑛𝑇); T is sampling interval

• Here, 𝑦 𝑛 = 𝑥 2𝑛 = 𝑥𝑎 (2𝑛𝑇);
Changes the sampling rate from 1/T to 1/2T
(i.e. decreasing rate by a factor of 2)

Why?
To reduce CPU time in a preliminary data analysis, or to reduce memory.
17 Department of CSE, CUET
 Manipulation of Discrete-Time Signals

 Addition, multiplication, and scaling of sequences:

 Amplitude modifications include addition, multiplication, and
scaling of discrete-time
 Amplitude scaling of a signal by a constant :

 Sum of two signals:

 Product of two signals:

18 Department of CSE, CUET

DISCRETE-TIME SYSTEM

19 Department of CSE, CUET

 Discrete-Time System

Definition: A discrete-time system is a device or algorithm

that operates on a discrete-time signal called the input or
excitation (e.g. 𝑥(𝑛)), according to some rule (e.g. 𝑇[.]) to
produce another discrete-time signal called the output or
response (e.g. 𝑦(𝑛)).

This expression denotes also the transformation 𝑇[.],(also

called operator or mapping) or processing performed by
the system on x(n) to produce y(n)

20 Department of CSE, CUET

 Classification of Discrete-Time Systems

 Static(Memoryless) Vs Dynamic system (Memory)

 Time varying Vs Time Invariant system
 Linear Vs Nonlinear System
 Causal Vs Noncausal System
 Stable vs. Unstable System

21 Department of CSE, CUET

Static(Memoryless) Vs Dynamic system (Memory)
 A discrete-time system is called static or memoryless if
its output at any time instant n depends on the input
sample at the same time, but not on the past or future
samples of the input.
 On the other hand, the system is said to be dynamic or
to have memory, if the output of a system at time n is
completely determined by the input samples in the
interval from n-N to n (𝑁 ≥ 0), the system is said to have
memory of duration N.
 If 𝑁 = 0, the system is static or memoryless.
 If 0 < 𝑁 < ∞ , the system is said to have finite memory.
 If 𝑁 → ∞, the system is said to have infinite memory.

22 Department of CSE, CUET

Static(Memoryless) Vs Dynamic system (Memory)

Examples:
 The static (memoryless) systems:

 The dynamic systems with finite memory:

 The dynamic system with infinite memory:

23 Department of CSE, CUET

 Time Varying Vs Time Invariant System

 A discrete-time system is called time-invariant if its

input-output characteristics do not change with time. In
the other case, the system is called time-variable.
 Time-Invariant (shift-invariant) Systems
 A time shift at the input signal x(n-n0) causes

corresponding time-shift at output signal y(n-n0)

24 Department of CSE, CUET

Time Varying Vs Time Invariant System

Example of Time-Invariant System

25 Department of CSE, CUET

 Time Varying Vs Time Invariant System

Example of Time-Varying System

 Memoryless System: 𝒚 𝒏 = 𝒏𝒙(𝒏)

26 Department of CSE, CUET

 Linear vs. Non-linear Systems

 A discrete-time system is called linear if only if it

satisfies the linear superposition principle. In the other
case, the system is called non-linear.

27 Department of CSE, CUET

Linear vs. Non-linear Systems

Example of Linear System

28 Department of CSE, CUET

 Linear vs. Non-linear Systems

Example of Nonlinear Systems

 Memoryless System: 𝒚 𝒏 = 𝒙𝟐 (𝒏)

29 Department of CSE, CUET

 Causal vs. Non-causal Systems
 A system is said to be causal if the output of the system
at any time n (i.e., y(n)) depends only on present and
past inputs (i.e., x(n), x(n-1), x(n-2), … ). In
mathematical terms, the output of a causal system
satisfies an equation of the form

where is some arbitrary function. If a system does not satisfy this

definition, it is called non-causal.
• Non-causal Systems are physically unrealizable for real-time
signal processing as can not observe future values
• Implementation is possible if processing of recorded signals done
offline

30 Department of CSE, CUET

 Stable vs. Unstable of Systems

 An arbitrary relaxed system is said to be bounded input -

bounded output (BIBO) stable if and only if every
bounded input produces the bounded output. It means,
that there exist some finite numbers say Mx and My,
such that

for all n. If for some bounded input sequence x(n) , the

output y(n) is unbounded (infinite), the system is
classified as unstable.

31 Department of CSE, CUET

Input-Output Description of Discrete-Time Systems

 Consists of a mathematical expression or a rule, which

explicitly defines the relation between the input and the
output signals
 Exact internal structure (i.e., black box) of the system is
either unknown or ignored

Graphical representation of a discrete-time signal

𝜏
𝑥(𝑛) 𝑦(𝑛)
Notation for general input-output relationship

32 Department of CSE, CUET

Block Diagram Description of Discrete-Time Systems
 The basic elements of discrete-time systems are the adder,
the multiplier, and the unit delay/ advance element
 Adder produces the sum of its inputs

 Multiplier
 Constant multiplier- multiplies its input by a constant
instantaneously
 Signal multiplier- multiplies two signal sequence to produce one

33 Department of CSE, CUET

Block Diagram Description of Discrete-Time Systems

 Unit delay element delays the signal passing through it

by one sample , i.e. 𝑦 𝑛 = 𝑥(𝑛 − 1)

 Unit advance element advances the signal passing

through it by one sample , i.e. 𝑦 𝑛 = 𝑥(𝑛 + 1)

34 Department of CSE, CUET

Block Diagram Description of Discrete-Time Systems

 Example
1 1 1
𝑦 𝑛 = 𝑦 𝑛−1 + 𝑥 𝑛 + 𝑥 𝑛−1
4 2 2
1 1
⇒ 𝑦 𝑛 = 𝑦(𝑛 − 1) + ,𝑥 𝑛 + 𝑥 𝑛 − 1 -
4 2
x(n) When n=n0,y(n-1)=0

0.5
x(n) + + y(n)

Z-1 x(n-1) Z-1

y(n-1)

0.25
Black Box

35 Department of CSE, CUET

 Interconnection of Discrete-Time Systems

 Can be interconnected to form large systems

 Two basic ways:
 In cascade (series)

 In parallel

36 Department of CSE, CUET

 Interconnection of Discrete-Time Systems
 Cascade interconnection

Output of first system: 𝑦1 𝑛 = 𝜏1 ,𝑥 𝑛 -

Output of second system: 𝑦 𝑛 = 𝜏2 𝑦1 𝑛
= 𝜏2 {𝜏1 ,𝑥 𝑛 -}
Overall system: 𝜏𝑐 ≡ 𝜏2 𝜏1
Output of overall system: 𝑦 𝑛 = 𝜏𝑐 ,𝑥 𝑛 -
 Order of performing 𝜏1 and 𝜏2 is important
 𝜏1 𝜏2 ≠ 𝜏2 𝜏1 for arbitrary system

 If 𝜏1 and 𝜏2 are linear and time invariant then

a) 𝜏𝑐 is time invariant
b) 𝜏1 𝜏2 = 𝜏2 𝜏1

37 Department of CSE, CUET

Interconnection of Discrete-Time Systems

38 Department of CSE, CUET

 Interconnection of Discrete-Time Systems

 To construct larger, more complex systems

 Conversely can take a larger system and break it down
into smaller subsystem for analysis and implementation

39 Department of CSE, CUET

ANALYSIS OF DISCRETE-TIME LINEAR
TIME-INVARIANT SYSTEMS

40 Department of CSE, CUET

 Techniques for Analysis of Linear Systems
Two basic methods for analyzing the behavior or response of a linear
system to a given input signal
 Method-1:
 Based on the direct solution of the input-output
equation for the system
 General form:

where F[.] is some arbitrary function.

A system whose output y(n) at time n depends on any number of the past
outputs values ( e.g. y(n-1), y(n-2), … ), is called a recursive system.
In contrast, if y(n) at time n depends only on the present and past inputs then
such a system is called Nonrecursive.

41 Department of CSE, CUET

Techniques for Analysis of Linear Systems (Method-1)

 A Linear Time Invariant (LTI) system can be described

by a linear constant coefficient difference equation of
the form:

Where, {ai} and {bi} are constant parameters that specify

the system and are independent of x(n) and y(n)

Difference Equation

42 Department of CSE, CUET

 Techniques for Analysis of Linear Systems

 Method-2:
 Decompose or resolve the input signal into sum of elementary
signals
Suppose,
 Input signal x(n) is resolved in to a weighted sum of

elementary signal components {xk(n)}

So that, 𝑥 𝑛 = 𝑘 𝑐𝑘𝑥𝑘 𝑛
Where, ck are the set of amplitudes in the decomposition of the signal x(n)
 Response of the system to the elementary signal component
xk(n) is yk(n)
Thus,𝑦𝑘(𝑛) ≡ 𝜏,𝑥𝑘 𝑛 -

43 Department of CSE, CUET

Techniques for Analysis of Linear Systems (Method-2)
 Assume
• system is relaxed (i.e. have zero initial state)
• response to ckxk(n) is ckyk
 Total response to the input x(n) is

𝑦 𝑛 =𝜏 𝑥 𝑛

=𝜏 𝑐𝑘𝑥𝑘 𝑛
𝑘

= 𝑐𝑘𝜏,𝑥𝑘 𝑛 -
𝑘

= 𝑐𝑘𝑦𝑘 𝑛
𝑘

44 Department of CSE, CUET

Resolution of Discrete-Time signal into impulse

Lets select the elementary signals xk(n) to be

𝑥𝑘 𝑛 = 𝛿(𝑛 − 𝑘)
where k represents the delay of unit sample

Thus, 𝑥 𝑛 𝛿 𝑛 − 𝑘 = 𝑥 𝑘 𝛿 𝑛 − 𝑘 is a sequence that

is zero everywhere except at n=k, where its value is x(k)
45 Department of CSE, CUET
 Resolution of Discrete-Time signal into impulse

 Repeating multiplication of x(n) with 𝛿(𝑛 − 𝑚) where

m is another delay (m ≠ 𝑘)
𝑥 𝑛 𝛿 𝑛−𝑚 =𝑥 𝑚 𝛿 𝑛−𝑚
 Hence, each multiplication of x(n) by a unit impulse at
some delay k, will results in a single value x(k) of the
signal x(n) at delay where the unit impulse is nonzero
 Repeating this multiplication over all possible delays,
− ∞ < 𝑘 < ∞ and adding all the product sequences will
result in a sequence equal to the sequence x(n) Summation of a n
Resolution/ ∞ infinite number of
decomposition of any scaled unit sample
arbitrary signal x(n)
into weighted sum of
𝑥 𝑛 = 𝑥(𝑘)𝛿(𝑛 − 𝑘) sequences where the
unit sample sequences
𝜹(𝒏−𝒌) has amplitude
shifted unit sample
sequence
𝑘=−∞ value of x(k)

46 Department of CSE, CUET

 Resolution of Discrete-Time signal into impulse

 Example

Consider a finite duration sequence

x(n)= {2, 4, 0, 3}
Resolve the sequence x(n) into sum of weighted impulse
sequences
Solution: Here, x(n) is nonzero at time instants, n=-1, 0, 2
∞
n -1 0 1 2
𝑥 𝑛 = 𝑥(𝑘)𝛿(𝑛 − 𝑘)
x(n) 2 4 0 3 𝑘=−∞

Need three impulses are delays, k=-1, 0, 2

Thus, 𝑥 𝑛 = 2𝛿 𝑛 + 1 + 4𝛿 𝑛 + 3𝛿(𝑛 − 2)

47 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum
Lets denote the response y(n,k) of the system to input unit sample
sequence n=k by special symbol h(n,k),
-∞ < 𝑘 < ∞
𝐼. 𝑒. , 𝑦 𝑛, 𝑘 ≡ ℎ 𝑛, 𝑘 = 𝜏,𝛿 𝑛 − 𝑘 -
Scaling the impulse at the input by an amount 𝑐𝑘 ≡ 𝑥(𝑘),
the response, 𝑐𝑘ℎ 𝑛, 𝑘 = 𝑥 𝑘 ℎ(𝑛, 𝑘)
Input arbitrary signal, 𝑥 𝑛 = ∞ 𝑘=−∞ 𝑥(𝑘)𝛿(𝑛 − 𝑘)
Then response , 𝑦 𝑛 = 𝜏 𝑥 𝑛
∞

= 𝜏, 𝑥(𝑘)𝛿(𝑛 − 𝑘)- Follows the

𝑘=−∞ superposition
∞ property on linear
= 𝑥 𝑘 𝜏 𝛿 𝑛−𝑘 systems and is
known as the
𝑘=−∞ superposition
∞
= 𝑘=−∞ 𝑥 𝑘 ℎ(𝑛, 𝑘) summation

48 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum

 If the system is time invariant, response of LTI

system to the unit sample sequence 𝛿(𝑛) is
denoted as h(n)
h(𝑛) ≡ 𝜏,𝛿 𝑛 -
According to time invariant property response to
the delayed sample sequence 𝛿 𝑛 − 𝑘 is
ℎ 𝑛 − 𝑘 = 𝜏,𝛿 𝑛 − 𝑘 -
∞
Hence, y(n)= 𝑘=−∞ 𝑥 𝑘 ℎ(𝑛 − 𝑘)

Convolution Sum
49 Department of CSE, CUET
Response to Arbitrary Inputs: The convolution Sum

Response as n=n0 is given as,

∞
y(n0)= 𝑘=−∞ 𝑥 𝑘 ℎ(𝑛0 − 𝑘)
 Computing the convolution between x(k) and h(k)
involves four steps:
 Folding

 Shifting

 Multiplication

 Summation

50 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum

Example:
Impulse response of a linear time variant system,
h(n)= {1, 2, 1, -1}
Determine the response of the system to the input signal,
x(n)= {1, 2, 3, 1}
Solution: Lets, illustrate impulse and input sequence-

Used k as index for keeping consistency with the equation of convolution sum

51 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum

 Compute convolution according to

∞
y(n)= 𝑘=−∞ 𝑥 𝑘 ℎ(𝑛 − 𝑘)
First step is to fold h(k),
i.e., h(-k)= {-1, 1, 2, 1}

52 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum

 Computing response at n=0,

y(0)= ∞ 𝑘=−∞ 𝑥 𝑘 ℎ(0 − 𝑘) = ∞
𝑘=−∞ 𝑥 𝑘 ℎ −𝑘
=2+2
=4
h(-k)= {-1, 1, 2, 1}
x(k)= {1, 2, 3, 1}
k -2 -1 0 1 2 3
h(-k) -1 1 2 1 0 0
x(k) 0 0 1 2 3 1
Here, product sequence, v0(𝑘) ≡ 𝑥 𝑘 ℎ(−𝑘)
k -2 -1 0 1 2 3
v0(k) 0 0 2 2 0 0

53 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum
 Computing response at n=1,
y(1)= ∞𝑘=−∞ 𝑥 𝑘 ℎ(1 − 𝑘)
=1+4+3
=8
Here, h(1-k) is folded sequence shifted to right

k -1 0 1 2 3 4
h(1-k) -1 1 2 1 0 0
x(k) 0 1 2 3 1 0
Here, product sequence, v1(𝑘) ≡ 𝑥 𝑘 ℎ(1 − 𝑘)
k -1 0 1 2 3 4
v1(k) 0 1 4 3 0 0

54 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum
 Similarly compute
y(2)= ∞
𝑘=−∞ 𝑥 𝑘 ℎ(2 − 𝑘) = 8
Hence,
y(3)= ∞
𝑘=−∞ 𝑥 𝑘 ℎ(3 − 𝑘) = 3 For n>5, y(n)=0
y(4)= ∞
𝑘=−∞ 𝑥 𝑘 ℎ(4 − 𝑘) = −2
y(5)= ∞
𝑘=−∞ 𝑥 𝑘 ℎ 5 − 𝑘 = −1
k -2 -1 0 1 2 3 4 5 6 7 8
h(-k) -1 1 2 1 0 0 0 0 0 0 0
h(1-k) 0 -1 1 2 1 0 0 0 0 0 0
h(2-k) 0 0 -1 1 2 1 0 0 0 0 0
h(3-k) 0 0 0 -1 1 2 1 0 0 0 0
h(4-k) 0 0 0 0 -1 1 2 1 0 0 0
h(5-k) 0 0 0 0 0 -1 1 2 1 0 0
h(6-k) 0 0 0 0 0 0 -1 1 2 1 0
x(k) 0 0 1 2 3 1 0 0 0 0 0

55 Department of CSE, CUET

Response to Arbitrary Inputs: The convolution Sum
 Computer y(n) for n<0
 Response at n=-1,
y(-1)= ∞𝑘=−∞ 𝑥 𝑘 ℎ(−1 − 𝑘)
=1
Here, h(-1-k) is folded sequence shifted to left

k -3 -2 -1 0 1 2 3
h(-1-k) -1 1 2 1 0 0 0
x(k) 0 0 0 1 2 3 1
Here, product sequence, 𝑣−1 (𝑘) ≡ 𝑥 𝑘 ℎ(1 − 𝑘)
k -1 0 1 2 3 4
𝒗−𝟏 (k) 0 1 0 0 0 0

56 Department of CSE, CUET

 Response to Arbitrary Inputs: The convolution Sum

 Any further shift to left will result in zero product

sequence
k -4 -3 -2 -1 0 1 2 3
h(-2-k) -1 1 2 1 0 0 0 0
x(k) 0 0 0 0 1 2 3 1

Hence, y(n)=0 for 𝒏 ≤ −𝟐

Respose of the system for ∞ < 𝒏 < ∞ is

y(n)={…, 0, 0, 1, 4, 8, 8, 3, -2, -1, 0, 0,…}

57 Department of CSE, CUET

 Properties of Convolution of LTI System

For convenient will use following notation to illustrate

convolution:

58 Department of CSE, CUET

 Properties of Convolution of LTI System

 Commutative Law

59 Department of CSE, CUET

 Properties of Convolution of LTI System

 Associative Law (Cascaded connection)

60 Department of CSE, CUET

 Properties of Convolution of LTI System

 Distributive Law (Parallel Connection)

61 Department of CSE, CUET

 Causal LTI Systems

 A relaxed LTI system is causal if and only if its impulse

response is zero for negative values of n , i.e.

Then, the two equivalent forms of the convolution formula

can be obtained for the causal LTI system:

62 Department of CSE, CUET

 Stable LTI Systems

 A LTI system is stable if its impulse response is

absolutely summable [i.e. every bounded input produce
a bounded output (BIBO)]

∞
𝑆ℎ ≡ 𝑘=−∞ ℎ(𝑘) < ∞

63 Department of CSE, CUET

Finite vs. Infinite-Duration LTI System Response

64 Department of CSE, CUET

Discrete Time Systems Described by Difference Equations

 FIR systems can easily be implemented by their

convolution summation (involves addition ,
multiplication, and a finite number of memory location)
 IIR systems, however, can not be practically
implemented by convolution by a infinite number of
memory location, multiplications, and additions.
 There is a practical and computationally efficient means
for implementing IIR systems through the use of
difference equation

65 Department of CSE, CUET

 Difference Equation

 An important subclass of LTI systems consist of those

system for which the input x(n) and output y(n) satisfy
an Nth-order linear constant-coefficient difference
equation.
 The general form:
N M
y ( n)   ak y ( n  k )   bk xn  k 
k 1 k 0

Or, N M

a
equivalently,
k 0
k y (n  k )  b
k 0
k x ( n  k ), a0  1

66 Department of CSE, CUET

 Solving the difference equation

 Goal is to determine the output y(n), n ≥ 0, of the

system given a specific input x(n), n ≥ 0, and a set of
initial conditions
 Direct Method

 Indirect Method- based on Z-transform

67 Department of CSE, CUET

 Direct Method

 Total solution is the sum of two parts:

𝑦 𝑛 = 𝑦ℎ (𝑛) + 𝑦𝑝 (𝑛)

homogenous or particular
complementary solution
solution

68 Department of CSE, CUET

 Direct Method
 The Homogeneous solution is response of the system to initial
condition
 Assuming that the input, x(n)=0. Since , this gives us the zero-
input response of the system N M

 a yn  k    b xn  k 
k 0
k
k 0
k
The zero-input response, which is what N
the system does with no input at all. This
is due to initial conditions, such as energy
stored in capacitors and inductors.
 a yn  k   0 When a =1
k 0
k 0

The solution is the form of an exponential: 𝑦ℎ 𝑛 = λ𝑛

𝑁 𝑛−𝑘
Substitute this in the previous equation: 𝑘=0 𝑎𝑘 λ =0
Or, λ𝑛 + 𝑎1 λ𝑛−1 + 𝑎2 λ𝑛−2 + ⋯ + 𝑎𝑁−1 λ + 𝑎𝑁 = 0
This is called characteristic polynomial of the system. It has N
roots and denotes by λ1 , λ2 , λ3 , …., λ𝑁

69 Department of CSE, CUET

 Direct Method (Homogeneous Solution)

The roots can be real or complex or some roots are

identical.
Let assume that roots are real and not identical, the
solution becomes
𝑦ℎ 𝑛 = 𝐶1 λ1𝑛 + 𝐶2 λ𝑛2 + ⋯ + 𝐶𝑁 λ𝑛𝑁
where, C1, C2, …, CN are weighting coefficients. These are
determined from the initial conditions.
If there are two identical roots, the solution becomes
𝑦ℎ 𝑛 = 𝐶1 λ1𝑛 + 𝐶2 𝑛λ𝑛2 + 𝐶3 𝑛2λ𝑛3 + ⋯ + 𝐶𝑚 𝑛𝑚−1 λ1𝑛
+ 𝐶𝑚+1 λ𝑛𝑚+1 + ⋯ + 𝐶𝑁 λ𝑛𝑁

70 Department of CSE, CUET

 Direct Method
 The Particular solution (zero state response): response to
applied input signal x(n), 𝑛 ≥ 0
Zero-state response, which is the output of
the system with all initial conditions zero

K is
scale
factor

71 Department of CSE, CUET

 Implementation of Discrete-Time Systems

 In general form, The difference equation is given by

N M
y (n)   ak y (n  k )   bk xn  k 
k 1 k 0
M
 The nonrecursive system is, v ( n)   b xn  k 
k 0
k

 The recursive system is,

N
y ( n)    ak y ( n  k )  v ( n)
k 1

72 Department of CSE, CUET

 Implementation of Discrete-Time Systems
N M
y (n)   ak y (n  k )   bk xn  k 
k 1 k 0

73 Department of CSE, CUET

 Implementation of Discrete-Time Systems
 As interchanging the order of the cascade linear time-invariant
systems, the overall system response remains the same

By reversing the order of these two systems, we obtain the form II structure
N M
w(n)   ak w(n  k )  x(n) y (n)   bk wn  k 
k 1 k 0

74 Department of CSE, CUET

Implementation of Discrete-Time Systems

Form I Structure

Form II Structure - Sometimes called “canonic

form”

75 Department of CSE, CUET

 Implementation of Discrete-Time Systems

 A special case of general case:

If some parameters like ak is set to 0, only nonrecurvise LTI
system will exist
M
y (n)   bk xn  k 
k 0
 The response of this system is said to be a weighted moving
average of the input signal
o Sometimes called a moving average (MA) system

FIR (Finite Impulse Response) system with an impulse response

76 Department of CSE, CUET

 Implementation of Discrete-Time Systems

 FIR systems can be realized from recursive and

nonrecursive systems
 Every FIR system can be realized nonrecursively
o Usual case
 FIR system can also be realized from recursive system
o moving average system

77 Department of CSE, CUET

 Correlation of Discrete-Time Signals

 Similar to convolution in the sense that two signal

sequences are involved
 Important to measure the degree of similarity between
two signals
 Widely used in radar, sonar, digital communications,
geology, and other areas in science and technology
 Example:
 Radar target detection

 Digital communication transmission

78 Department of CSE, CUET

 Crosscorrelation and Autocorrelation Sequence

 The crosscorelation of x(n) and y(n) is a sequence rxy(l)

is defined as
Here,
• l is the (time) shift
(or lag) parameter
• Subscript xy
The reverse crosscorrelation is indicate sequence
being co related

79 Department of CSE, CUET

 Crosscorrelation and Autocorrelation Sequence

 In special case , we have autocorrelation, which is

defined as So, that

80 Department of CSE, CUET

 Summary

 Characterization of discrete –time signals and systems in

time domain
 Linear Time-Invariant (LTI) System
 Responses

 Properties

 Classifications

 Implementation

 Correlation of Discrete-Time Signals

81 Department of CSE, CUET

 Book

 John G. Proakis, Dimitris G. Manolakis, “Digital Signal

Processing: Principles, Algorithms, and Applications”,
Prentice Hall, Fourth Edition
Practice exercises and problems of the Chapter

82 Department of CSE, CUET

Questions or comments?

83 Department of CSE, CUET