You are on page 1of 184

Introduction

Prof.Hansa Shingrakhia

Signal, System, Signal


Processing
Signal
A signal is a function of independent variables such as
time, distance, position, temperature and pressure
Signals play an important role in our daily life.
Most signals we encounter are generated naturally.
However, a signal can also be generated synthetically or by
a computer.
Depending on the nature of the independent variables and
the value of the function, we can classify the signals as
follows

Signal, System, Signal


Processing
scalar signal
A signal generated by a single source
vector signal (or multichannel signal)
A signal generated by multiple sources
one dimensional (1-D) signal
A function of a single independent variable
multidimensional (M-D) signal
A function of more than one independent variables
sound

Image

Video

Signal, System, Signal


Processing
continuous-time signal

x(t )

The independent variable is continuous

discrete-time signal (sequence)

see

x(n)

The independent variable is discrete

analog signal
A continuous-time signal with a continuous amplitude

quantized-boxcar signal
A continuous-time signal with a discrete-value amplitude

digital signal

A discrete-time signal with a discrete-value amplitude

Sampled-data signal
A discrete-time signal with a continuous amplitude

Signal, System, Signal


Processing
deterministic signal
A signal that can be uniquely determined by a well-defined
process, such as a mathematical expression or rule, or table
look-up.

random signal
A single that is generated in a random fashion and cannot be
predicted ahead of time.

Signal, System, Signal


Processing
Periodic signal :A signal that repeats itself in a periodic
fashion from negative to positive infinity.

x ( t ) x ( t kT )
x ( n) x ( n kN )
Aperiodic signal: A signal that extends to both positive and
negative infinity without repeating in a periodic pattern.

Signal, System, Signal


Processing
power signal
An infinite energy signal with finite average power.

1
P lim
T T

x ( t ) dt

1
P lim
N N

N 1

x ( n)

n 0

energy signal
An finite energy signal with zero average power.

x ( t ) dt

E x ( n)

a periodic sequence power signal


a finite-length sequence energy signal

Signal, System, Signal


Processing
System
A system is any process that produces an output signal
in response to an input signal.
Depending on the types of the signal processed, we
can classify the systems as follows:

Signal, System, Signal


Processing
analog system
input and output analog signals
digital system
input and output digital signals
continuous-time system
input and output continuous-time signals
discrete-time system
input and output discrete-time signals

Signal, System, Signal


Processing
Signal Processing
A signal carries information !
The objective of signal processing:
To extract, enhance, store and transmit the useful information
carried by the signal.
Digital signal processing:
To implement the signal processing by a digital means.
return

The Constitution of DSP


System
Equivalent to an analog signal processor
analog

analog
PrF

ADC

DSP

DAC

PoF

waveform

PrF: antialiasing filtering


PoF: smooth out the staircase waveform
return

example

How to Implement the DSP

To handle the DSP algorithms in a general-purpose microprocessor


by means of software programming
To handle the DSP algorithms in a specifically designed Digital
Signal Processors (DSPs)

return

The Application of DSP


The main tasks of DSP

Signal Analysis

Measurement of signal properties


Spectrum (frequency/phase) analysis
Target detection, verification, recognition

Signal Filtering

Signal-in-signal-out, filter
Removal of noise/interference
Separation of frequency bands

The Application of DSP


DSP application examples

Telecommunications

Multiplexing
Compression
Echo control

Audio Processing

Music
Speech generation
Speech recognition

The Application of DSP


DSP application examples

Echo Location

Radar
Sonar
Reflection seismology

Image Processing

Medical
Space
Commercial Imaging Products

The Application of DSP


Demonstration
FM synthesis of sound
chirp

clarinet

brass

bell

Digital image processing


Deblurring
Edge detection
Noise reduction
return

The Advantages of DSP

Absence of drift in the filter characteristics


Improved quality level
Reproducibility
Ease of new function development
Multiplexing
Modularity
Total single chip implementation using VLSI technology
return

The sound signal is an example of a 1-D signal


where the independent variable is time

return

A black-and-white image signal is an example of a


2-D signal where the 2 independent variables are the
2 spatial variables.

I ( x, y)

A color image signal is a 3-channel signal composed of three


2-D signals representing the three primary color: red, green
and blue (RGB)

I R ( x , y )
u( x , y ) I G ( x , y )
I G ( x , y )

return

A black-and-white video signal is an example of a 3-D signal


where the 3 independent variables are the 2 spatial variables
and the time variable.

I ( x, y, t )

A color video signal is a 3-channel signal composed of three


3-D signals representing the three primary color: red, green
and blue (RGB)

I R ( x , y , t )

u( x , y , t ) I G ( x , y , t )
I B ( x , y , t )

return

1 1
0.9
1
1 0.9
0.8 0.8
0.9
0.9
0.7 0.7
0.8
0.8
0.6 0.6
0.7
0.7
0.5
0.5
0.6 0.6
0.4 0.4
0.5
0.5
0.3
0.3
0.4 0.4
0.2 0.2
0.3
0.3
0.1
0.1
0.2 0.2
0
0.10 0.1
-0.1
-0.1
0 0
-0.2 -0.2
-0.1
-0.1
-0.3
-0.2
-0.2 -0.3
-0.4 -0.4
-0.3
-0.3
-0.5
-0.5
-0.4 0
-0.4 0
-0.5 -0.5
0 0
1 1
0.9 0.9
1
1 0.8
0.8
0.9
0.9
0.7 0.7
0.8
0.8
0.6 0.6
0.7 0.7
0.5 0.5
0.6
0.6
0.4 0.4
0.5
0.5
0.3 0.3
0.4
0.4
0.2 0.2
0.3
0.3
0.1 0.1
0.20 0.2
0
0.1
0.1 -0.1
-0.1
0
0 -0.2
-0.2
-0.1
-0.3 -0.1
-0.3
-0.2
-0.4 -0.2
-0.4
-0.3 -0.3
-0.4 0
-0.4 0

10 10
10 10

10 10

20 20
20 20

20 20

30 30
30 30

1 1
0.9
1 0.91
0.8 0.8
0.9
0.9
0.7
0.7
0.8 0.8
0.6 0.6
0.7
0.7
0.5
0.5
0.6 0.6
0.4 0.4
0.5
0.5
0.3
0.3
0.4 0.4
0.2 0.2
0.3
0.3
0.1
0.1
0.2 0.2
0
0.10 0.1
-0.1
-0.1
0 0
-0.2 -0.2
-0.1
-0.1
-0.3
-0.3
-0.2 -0.2
-0.4 -0.4
-0.3
-0.3
-0.5
-0.5
-0.4 0-0.40
-0.5 -0.5
0 0

10 10
10 10

20 20
20 20

30 30

1 1
0.9 0.9
1 0.81
0.8
0.9
0.9
0.7 0.7
0.8
0.8
0.6 0.6
0.7 0.7
0.5 0.5
0.6
0.6
0.4 0.4
0.5
0.5
0.3 0.3
0.4
0.4
0.2 0.2
0.3
0.3
0.1 0.1
0.20 0.2
0
0.1
0.1 -0.1
-0.1
0 -0.20
-0.2
-0.1
-0.1
-0.3 -0.3
-0.2
-0.2
-0.4 -0.4
-0.3 -0.3
-0.4 0-0.40

10 10

20 20

30 30
30 30

30 30

return

10

20

30

40

50

10

20

30

40

50

10

20

30

40

50

10

20

30

40

50

10

20

30

40

50

10

20

30

40

50
return

Original Image

Restored Image

Blurred Image

return

return

return

Discrete-Time Signals and


Systems
Content
The Discrete-Time Signal: Sequence
The Discrete-Time System
The Discrete-Time Fourier Transform (DTFT)
The Symmetric Properties of the DTFT
System Function and Frequency Response

The Discrete-Time Signal:


Sequences
Elementary sequences

Unit sample sequence

1, n 0
( n)
0, n 0

1, n n0
( n n0 )
0, n n0

The Discrete-Time Signal:


Sequences

Unit step sequence

1, n 0
u( n)
0, n 0

( n) u( n) u( n 1)

u( n) ( n m )
m 0

1, n n0
u( n n0 )
0, n n0

The Discrete-Time Signal:


Sequences

Rectangular sequence

1, 0 n N 1
RN ( n )
otherwise
0,
RN ( n) u( n) u( n N )
N 1

RN ( n) ( n m )
m 0

The Discrete-Time Signal:


Sequences

Sinusoidal sequence

x ( n) A cos( 0 n ),
A
0

amplitude
digital angular frequency
phase

x1 ( n) 1.5 cos(0.05 2 n)
x 2 ( n) 1.5 sin( 0.05 2 n)

The Discrete-Time Signal:


Sequences

Real-valued exponential sequence

x ( n) a , n; a R
n

The

x (n) is convergent when | a | 1

The

x (n) is divergent when | a | 1

x1 ( n) 0.001 1.2
x 2 ( n) 0.2 0.8 n

The Discrete-Time Signal:


Sequences

Complex-valued exponential sequence

x ( n) e

( j 0 ) n

n
n

x( n) e cos 0 n je sin 0 n
x re ( n) jx im ( n)

Attenuation factor

x ( n ) 2e

1
( j )n
5 8

The Discrete-Time Signal:


Sequences
Classification of sequences
Finite-length sequence

x(n) is defined only for a finite time interval:


N1 n N 2
examples

where N 1 , N 2

x ( n) n 2 ,
8 n 8
y( n) cos 0.4n

The length of a finite-length sequence can be


increased by zero-padding

The Discrete-Time Signal:


Sequences

Right-sided sequence

x(n) has zero-valued samples for n N 1


where N 1
If N 1 0 , a right-sided sequence is called a
causal sequence

The Discrete-Time Signal:


Sequences

Left-sided sequence

x(n) has zero-valued samples for n N 2


where N 2
If N 2 0 , a left-sided sequence is called a
anti-causal sequence

The Discrete-Time Signal:


Sequences

Two-sided sequence

x (n) is defined for any n


a dual-sided sequence can be seen as the
sum of a right-sided sequence and a left-sided
sequence.

The Discrete-Time Signal:


Sequences

Absolutely summable sequence

x ( n)

Example:

0.3 n ,
x( n)
0,

1
0.3
1.42857

1 0.3
n0
n

n0
n0

The Discrete-Time Signal:


Sequences

Square-summable sequence

x( n)

Example:

sin 0.3n
x ( n)
n

It is square-summable but not


absolutely summable

The Discrete-Time Signal:


Sequences
Operations on sequence

Time-shifting operation

y( n) x ( n N )

where N is an integer

N 0 delaying operation
Unit delay

x (n)

z-1

y( n) x ( n 1)

y( n) x ( n 1)

N 0 advance operation
Unit advance x (n)

Sequence shifts
Sequence x(n)

Delayed sequence

x(n-3)

Advanced sequence

x(n+2)

The Discrete-Time Signal:


Sequences

Time-reversal (folding) operation

y( n) x ( n)

Addition operation
Sample-by-sample addition
Adder

y ( n) x ( n) w ( n)
y ( n) x ( n) w ( n)

x (n)
w (n)

Examples
2

Find the following


1

1.x(n) + y(n)
2.x(n) . y(n)

x(n)
0 1 2 3 4 5 6 7
2.0
1.5

0.5

1.0

1.0

4.x(n-3)
5.y(n+2)

0.5

y(n)

3
n 0 1 2

3.2x(n) + 3y(n)

7
4 5 6

6.x(-n)

-0.5
-2.0

Graph, Table, Sequence

The Discrete-Time Signal:


Sequences

Scaling operation

y( n) Ax ( n)
Multiplier

x (n)

y( n) Ax ( n)

Product (modulation) operation


Sample-by-sample multiplication y( n) x ( n) w ( n)

modulator

y ( n) x ( n) w ( n)

x(n)
w (n)

The Discrete-Time Signal:


Sequences

Sample summation
n2

x ( n) x( n ) x ( n )

n n1

Sample production
n2

x ( n) x ( n ) x ( n )

n n1

The Discrete-Time Signal:


Sequences

Sequence energy

Ex

*
2
x
(
n
)
x
(
n
)

|
x
(
n
)
|

Sequence power

1
Px lim
N N

N 1

| x ( n) |
n 0

The Discrete-Time Signal:


Sequences
Sequence synthesis

Unit sample synthesis

Any arbitrary sequence can be synthesized in the timedomain as a weighted sum of delayed (advanced) and
scaled unit sample sequence.

x ( n)

x(k ) (n k )

The Discrete-Time Signal:


Sequences

Even and odd synthesis

Even (symmetric): xe ( n) xe ( n)
Odd (antisymmetric): xo ( n) xo ( n)
Any arbitrary real-valued sequence can be
decomposed into its even and odd component:

x ( n) xe ( n) xo ( n)

1
xe ( n) [ x( n) x( n)]
2
1
xo ( n) [ x( n) x ( n)]
2

return

The Discrete-Time System


Introduction
A discrete-time system processes a given input
sequence x(n) to generate an output sequence y(n)
with more desirable properties.
Mathematically, an operation T [ ] is used.
y(n) = T [ x(n) ]
x(n): excitation, input signal
y(n): response, output signal

example
example

The Discrete-Time System


Classification

Memoryless (static) system

Linear System

Time-Invariant (Shift-Invariant) System

Linear Time-Invariant (LTI) System

Causal System

Stable System

The Discrete-Time System


Linear System
A system is called linear if it has two mathematical
properties: homogeneity and additivity.

T [ax ( n)] aT [ x( n)]


T [ x1 ( n) x 2 ( n)] T [ x1 ( n)] T [ x 2 ( n)]

T [a1 x1 ( n) a 2 x 2 ( n)] a1T [ x1 ( n)] a 2T [ x 2 ( n)]


Accumulator

The Discrete-Time System


Time-Invariant (Shift-Invariant) System

if T [ x ( n)] y( n)
then T [ x ( n n0 )] y( n n0 )
Accumulator

The Discrete-Time System


Linear Time-Invariant (LTI) System
A system satisfying both the linearity and the timeinvariance properties is called an LTI system.
LTI systems are mathematically easy to analyze and
characterize, and consequently, easy to design.
A accumulator is an LTI system !

The Discrete-Time System


The output of an LTI system is called
linear convolution sum

y( n) LTI [ x ( n)]

x(k )h(n k ) x(n) * h(n)

An LTI system is completely characterized in the time


domain by the impulse response h(n).

example

The Discrete-Time System


Causal System
Output depends only on current and past inputs (output is not
anticipative)
y(n)=x(n+1)-x(n) Non-causal (forward difference)
y(n)=x(n)-x(n-1) Causal
(backward difference)

The Discrete-Time System


An LTI system will be a causal system if and only
if :

h( n) 0,

n0

An ideal low-pass filter is not a causal system !


A sequence

x (n) is called a causal sequence if :


x ( n ) 0,

n0

The Discrete-Time System


Stable System
A system is said to be bounded-input bounded-output
(BIBO) stable if every bounded input produces a
bounded output, i.e.

if

x ( n) M , then y( n) P

An LTI system will be a stable system if and only if :

h(n)

Linear system - scaling


Input
x(n)

Output
y(n)

Impulse
Response

2x(n)

2y(n)

3x(n)

3y(n)
x(n)

y(n)

Linear system
-superposition
x(0)h(n-0)

x(n)
Input

x(1)h(n-1)

Impulse
Response

x(2)h(n-2)

y(n) = x(k)h(n-k)

Output

Superposition of responses
Linearity allows
scaling of impulse
response for samples
of different amplitude
Superposition allows
adding responses of
delayed input
samples
Result is the
convolution of input
sequence and system
impulse response

Examples
y(n)
1. Cos[x(n)]
2

n 1

x(k )

3. x(n)cos(0n)
4. x(-n+2)
5. Quant[x(n)]
6. x(n)+nx(n+1)

Static/
Linear/
Dynamic Nonlinear

Time
invariant/
time
varying

Causal/
Noncausal

Stable/
Unstable

Examples
Static/
Linear/
Dynamic Nonline
ar

Time
invaria
nt/ time
varying

Causal
/Noncausal

Stable/
Unstab
le

NL

TI

x(k )

TI

NC

US

3. x(n)cos(0n)

TV

4. x(-n+2)

TI

NC

TV

NC

US

y(n)
1. Cos[x(n)]
2.

n 1

x(n)+nx(n+1)

Systems that are linear and whose response does not


change with time (time independent)
If y(n) = T[x(n)], then y(n-n0)=T[x(n-n0)]

LTI is based on the principle of super-position

Linear system can be completely characterized by its


impulse response

If hk(n) is the response of the system to an impulse (n-k)


occurring at n=k then the output the system for an input
sequence x(n)=x(k)(n-k) (k=- to ) is

y ( n) T

x(k ) (n k )

x(k )T (n k ) x(k )h (n)

h1(n)
y(n)

x(n)

x(n)

h(n)=h1(n)+h2(n)

y(n)

h2(n)

x(n)

h1(n)
Cascaded

h2(n)

y(n)

x(n)

h1(n)*h2(n)

y(n)

x(n)

Z-1

x(n)
h(0)

h(1)

Z-1

Z-1

x(n-1)

x(n-2)

X h(2)

h(k-1)

Z-1

x(n-k)

x(n-k-1)
X

h(k)

X
y(n)

N 1

y ( n) h( k ) x ( n k )
k 0

Convolution Sum
R.N.Mutagi

Adders
Multiplie
rs
Memory

Involves only current and past input values

n
x(n
)
h(
n) 2

1
h(n)

2
1

4
2
1

5
4
2
1

5 2
4 5 2
2
1

4
2
1

5
4
2
1

y(n)

2
5
4
2
1

2
5
4
2

Rotate h(n) around 0 and


y(n
)
2 shift
9 22 40 46 30 14 5

2
5
4

2
5

2
5
4
2
1

1
2
5
4
2
1

x(n)
2
4
4
8
10 20
8 16
4
8
2
4

5
10
25
20
10
5

Difference equation

yn a yn k bm xn m

(1)

yn is a function of ak, bm, xn, memory = states.

or yn hk xn k h0 xn hk xn k

(2)

If we know the initial conditions, we can solve the difference equation:

x1,n y1, n
predictionsof system ' sbehavior
x2,n y2, n

If (1) is a linear constant coefficient difference equation, the


output consists of a homogeneous solution and a particular
solution: yn = yh,n + yp,n
(2.3.3)

69

Homogeneous (complementary) solution


xn 0,n 0theoutputisduetotheinitialconditions
N

Homogeneous equation:

a
k 0

yn k 0

Let ' sassume yn ak n k 0


n

k 0

Then n N (a0 N a1 N 1 K an 1 an ) 0is a characteristicequation


The characteristic equation has N characteristic roots: 1, N
If the system is real, xn, yn, and all the coefficients (a, b) are real
hn are real roots are either real or complex conjugate pairs
For LTI systems
N
yh , n ci in
is a form of solution, where ci depend on initial conditions.

i 1
if 1 = 2, solution in form of
ELEN 5346/4304 DSP and Filter Design

Fall 2008

yh ,n c11n c2 n1n

Particular solution
We apply an input and see what happens to the output for large n
Assumption: yp,n has the same form as xn:

xn

yp,n

An M

k0 n M k1n M 1 K k M

cos 0 norsin 0 n

k1 cos 0 n k2 sin 0 n

cn

knc n

An n M
Ex.:

An (k0 n M k1n M 1 K k M )

5
1
1
yn 1 yn 2 xn xn 1 y1 6; y2 6; x1 1; xn 2n ,n 0
6
6
2
Characteristic equation:
yn

(2.5.1)

5
1
1
1
1

1
1
1
0 01 ;2 yh, n c1 c2 ;n 0 formof solution
6
6
2
2
3

3
2
3
(2.5.2)
2

Example (cont)
yn

5
1
1
yn 1 yn 2 xn xn 1 y1 6; y2 6;x1 1; xn 2n ,n 0
6
6
2
n

Got so far:

yh , n

is LTI
n

1
1
c1 c2 ;n 0
2
3

y zi ,n

1
1
7.5 3 ;n 0
2
3

Sincexn c n 2n y p ,n kc n k 2n
5
1
1
10
1

k 2n k 2n 1 k 2n 2 2n 2 n 12n 2 4k k 2n 2 (4 1)
6
6
2
6
6

1
n
4

1
k

2
p ,n

n
1

1
Totalsolution : yn yh ,n y p ,n c1 c2
2
3

2 2n

72

4. Correlation

72
ELEN 5346/4304 DSP and Filter Design

Fall 2008

73

Correlation and convolution

73
ELEN 5346/4304 DSP and Filter Design

Fall 2008

74

Autocorrelation

74
ELEN 5346/4304 DSP and Filter Design

Fall 2008

75

Correlation maxima

75
ELEN 5346/4304 DSP and Filter Design

Fall 2008

x(n)

y(n)=F{x(n),
y(n)=F{x(n),
x(n-1),..,
x(n-1),.., x(n-M)}

y(n)

x(n)

y(n)=F{y(n-N), ..,y(n-1),
..,y(n-1),
x(n), x(n-1),.., x(n-M
x(n-M)}
Delay

In a non-recursive system the output depends only on


the current and past input samples
Example: y(n)=x(n)+x(n-2)+x(n-5)

In a recursive system the output depends both on past


and current input samples and past output samples
Example: y(n)=x(n)+x(n-2)+y(n-3)+y(n-1)

y(n)

Cumulative average of x(n) in the interval 0<k<n is defined as

1 N
y ( n)
x(k ),

n 1 k 0

n 0,1,2,...

Computing y(n) non-recursively requires increasing memory. It


can be computed recursively as
N

(n 1) y (n) x(k ),

n 0 ,1,2 ,...

k 0

N 1

x ( k ) x ( n)

y(n)

x(n)

1/(n+1)

k 0

ny (n 1) x(n)
n
n
y ( n)
y (n 1)
x ( n)
n 1
n 1

Z-1
n

Fibonacci sequence is generated recursively

y(k+2)=y(k) + y(k+1)

Y(k+2) Y(k+1)
D

Y(k)

1
x ( n)
y (n) y (n 1)
2
y (n 1)
y(n) is the square root on nth iteration
y(n-1 ) is the initial estimate of square root (guess)
x(n) is the number whose square root is sought
n

y(n)

y(n-1)

x(n)/y(n-1)

3/3=1

1.75

3/2=1.5

1.732142857 1.75

3/1.75=1.714

1.7320

3/1732143=1.732

1.732142857

Eigen functions
If a linear operator T transforms an input function x(n) in to

the same function at the output with some scaling then


the function is eigen function and the scale factor is eigen
value

T [ x(n)] x(n)
d j t
Example : e e jt
dt
Complex exponential sequences are eigenfunctions of
LTI systems and the response to a sinusoidal input is a
sinusoidal with same frequency whose amplitude abd
phase are determined by the system

h( k )
j

H (e )

jk
h
[
k
]
e

d/dt

x ( n) e

j n

y ( n)

j ( n k )
h
(
k
)
e

jn

h[k ]e

k
H e j e j n
Eigenvalue H (e j ) is the frequency response

jk

Frequency response of a discrete-time LTI system is always a


periodic function of frequency variable with period 2

Periodic
frequency
response of ideal
lowpass filter

One period of the


frequency response
of ideal lowpass filter

Periodic ideal
highpass filter

Periodic ideal band


reject filter

Periodic ideal badpass


filter

Sequences can be represented by Fourier integral of the


form

1
j
jn
x[n]
X
(
e
)
e
d Synthesis equation IFT

2 coefficient discrete frequency

x(n) is a superposition of infinitesimally small complex


sinusoids of frequency ejn and amplitude X(ej)
The coefficients are obtained by

X (e )

jn
x
[
n
]
e

Analysis equation FT

Coefficients obtained from sequence x[n] determine how


much each frequency component be added to get x[n]

Frequency response of a linear time-invariant


system is the Fourier transform of the impulse
response of the system
Impulse response can be obtained from the
frequency response by using inverse
Fourier

1
j
jn
H (e j ) h[n]e jn
transform
h[n]
H
(
e
)
e
d

IFT

FT
Periodic frequency
response

Impulse
response

Conjugate symmetric sequence xe[n] is a sequence for which

xe [n] xe[n]

Conjugate anti-symmetric sequence xo[n] is a sequence for


which
x [ n] x [ n]
o

Any sequence x [n] can be expressed as

x[n] xe [n] xo [n]

where

1
xe [n] x[n] x [n] and
2
1
xo [n] x[n] x [ n]
2

A real conjugate
symmetric sequence is
called even sequence

xe [n] xe [ n]

A real conjugate antisymmetric sequence is


called odd sequence

xo [n] xo [ n]

x[n]

0
x[n]

A Fourier transform X(ej)can be decomposed in to


conjugate symmetric and conjugate anti-symmetric
functions

X ( e j ) X e ( e j ) X o ( e j )

1
where X e (e ) X (e j ) X (e j )
2
1
j
X e ( e ) X ( e j ) X ( e j )
2
j

X e e j X e e j

and

X o e j X o e j

Sequence x[n]

Fourier transform X(ej)

x*[n]

X*(e-j)

x*[-n]

X*(ej)

Re {x [n]}

Xe (ej)

Im {x [n]}

Xo (ej) conjugate anti-symmetric part of X(ej)

xe[n] conj.symm. part of x[n]

XR (ej)

xo[n] conj.antisymm. part of x[n]

jXI (ej)

Any real x [n]

X (ej)=X*(e-j) FT is conjugate symmetric

Any real x [n]

XR (ej)=XR (e-j)

Real part is even

Any real x [n]

XI (ej)=-XI (e-j)

Imaginary part is odd

Any real x [n]

|X (ej)|=|X (e-j)|

Any real x [n]

arg X (ej)=- arg X (e-j)

xe[n]

XR (ej)

even part of x[n]

conjugate symmetric part of X(ej)

Magnitude is even
Phase is odd

The Discrete-Time System


The M-point moving average filter is BIBO
stable :

1
y( n)
M

M 1

x( n k )
k 0

prove

A causal LTI discrete-time system:

h( n) n u( n)
prove

if | | 1, the system is stable


if | | 1, the system is not stable

The Discrete-Time System


Causal and Stable System
A system is said to be a causal and stable system if
the impulse response h(n) is causal and absolutely
summable , i.e.

h( n) h( n)u( n)

h(n)

return

The Discrete-time Fourier Transform


(DTFT)
The transform-domain representation of
discrete-time signal
Discrete-Time Fourier Transorm (DTFT)
Discrete-Fourier Transform (DFT)
z-Transform

The Discrete-time Fourier Transform


(DTFT)
The definition of DTFT
DTFT: X ( e ) DTFT [ x ( n)]
jw

jwn
x
(
n
)
e

1
IDTFT: x ( n) IDTFT [ X ( e )]
2
jw

Existence condition:

| x ( n) |

X ( e jw )e jwndw

The Discrete-time Fourier Transform


(DTFT)
The comparison of x (n) vs. X (e j )
x (n)

X (e j )

Time domain

Frequency domain

discrete

continuous

Real valued

Complex-valued

Summation

integral

The Discrete-time Fourier Transform


(DTFT)
About X (e j )
It is a periodic function of with a period of 2

~
The integral range of ~
The range of

It can be expressed as

X ( e j ) X ( e j ) e j ( )

X ( e j )

magnitude function

( )

phase function

X (e j ) and ( ) are all real function of

example

The Discrete-time Fourier Transform


(DTFT)
DTFT vs. z Transform

X ( e ) X ( z ) z e j

x ( n)e

j n

The Discrete-time Fourier Transform


(DTFT)
The general properties of DTFT
Linearity

The DTFT is a linear transformation

ax ( n) by( n) aX (e j ) bY (e j )
Time

shifting
A shift in the time domain corresponds to the
phase shifting

x( n m ) e

j m

X (e )

The Discrete-time Fourier Transform


(DTFT)
Frequency

shifting

Multiplication by a complex exponential


corresponds to a shift in the frequency domain

jn 0

x ( n) X (e

j ( 0 )

Convolution

Convolution in time domain corresponds to


multiplication in frequency domain

x1 ( n) x 2 ( n) X 1 (e ) X 2 (e )
jw

jw

The Discrete-time Fourier Transform


(DTFT)
Multiplication

1
x1 ( n) x 2 ( n)
2
Energy

X (e

) X 2 (e

(Parsevals Theorem)

1
x ( n)

2
n
2

X (e )

X (e ) d

energy density spectrum

j ( w )

)d

The Discrete-time Fourier Transform


(DTFT)
Multiplied

by an exponential sequence

1 j
a x ( n) X ( e )
a
n

Sequence

weighting

d
j
n x ( n) j
[ X ( e )]
d

The Discrete-time Fourier Transform


(DTFT)
Conjugation

Conjugation in the time domain corresponds to the


folding and conjugation in the frequency domain

x ( n ) X ( e j )

Folding

Folding in the time domain corresponds to the folding


in the frequency domain
j

x ( n) X (e

Conjugation

and Folding
Conjugation and folding in the time domain corresponds
to the conjugation in the frequency domain

x ( n ) X ( e j )

return

The symmetric properties of


the DTFT
Conjugate symmetry of x (n)

Conjugate symmetric sequence:

xe ( n) xe ( n)

For real-valued sequence, it is even symmetric:

xe ( n) xe ( n)

Conjugate antisymmetric sequence: x

( n) xo ( n)

For real-valued sequence, it is odd symmetric:

xo ( n) xo ( n)

The symmetric properties of


the DTFT
Any arbitrary sequence can be expressed as
the sum of a conjugate symmetric sequence and
a conjugate antisymmetric sequence

x ( n) xe ( n) xo ( n)
1
xe ( n) [ x( n) x ( n)]
2
1
xo ( n) [ x ( n) x ( n)]
2

The symmetric properties of


the DTFT
Conjugate symmetry of X ( e j )
The X (e j ) can be expressed as the sum of the
conjugate symmetric component and the conjugate
antisymmetric component

X ( e j ) X e ( e j ) X o ( e j )
1
X e (e ) [ X (e j ) X (e j )]
2
1
j
X o (e ) [ X (e j ) X (e j )]
2
j

The symmetric properties of


the DTFT
conjugate symmetric

X e ( e ) X e ( e j )

For real-valued function, it is even symmetric

X e ( e j ) X e ( e j )
conjugate antisymmetric

X o ( e ) X o ( e j )

For real-valued function, it is odd symmetric

X o ( e j ) X o ( e j )

The symmetric properties of


the DTFT
The symmetric properties of the DTFT

Re[ x ( n)]
j Im[ x ( n)]

X e (e )
j

X o (e )
j

xe ( n)

Re[ X (e )]

xo ( n)

j Im[ X (e )]

Implication: If the sequence x (n) is real and even,


then X (e j ) is also real and even.

The symmetric properties of


the DTFT
If the sequence x(n) is real, then
j

X (e ) X (e

Re[ X (e )] Re[ X (e
j

Im[ X (e )] Im[ X (e
j

X (e ) X (e
j

)]

)]

arg[ X (e )] arg[ X (e

)]

example

The symmetric properties of


the DTFT
The DTFT of periodic sequences
The

DTFT of complex-valued exponential sequences

x ( n) e
X (e

j 0 n

( n )

2 (

( 0 )

2i )

The symmetric properties of


the DTFT
The

DTFT of constant-value sequences

x ( n) 1
j

X (e )

(- n )

2 ( 2i )

The symmetric properties of


the DTFT
The

DTFT of unit sample sequences

x ( n)

(n iN )

2
X (e )
N
j

2
(
k)

N
k

The symmetric properties of


the DTFT
The

x~ ( n)

DTFT of general periodic sequences

x( n iN ) x( n) ( n iN )

2
X (e )
N
j

X (e

2
k
N

2
) (
k)
N

2
~
X ( k ) (
k)

N
k

return

System Function and Frequency


Response
The representation of a LTI system

Impulse response h(n)


N

Difference equation

a
k 0

System function H (z )

y( n k ) bm x ( n m )
m 0

System Function and Frequency


Response
System function (Transfer function)
The z-transform of the impulse response h(n) of the LTI
system is called system function or transfer function

H ( z ) [h( n)]

h(n)z

H (z) Y (z) X (z)

System Function and Frequency


Response
The region of convergence (ROC) for H(z)
An LTI system is stable if and only if the unit circle is in
the ROC of H(z)
An LTI system is causal if and only if the ROC of H(z) is

Rx | z |
An LTI system is both stable and causal if and only if
the H(z) has all its poles inside the unit circle, i.e. the
ROC of H(z) is

1| z |

System Function and Frequency


Response
System function vs. difference equation
N

difference equation

a
k 0

y( n k ) bm x ( n m )
m 0

take z-transform for both sides

k 0

Y (z)
H (z)

X (z)

m 0
N

k
a
z
k
k 0

m 0

m
b
z
m

m
a
z
Y
(
z
)

b
z
k
m X (z)
k

1
(
1

c
z
)

m
m 1
N

1
(
1

d
z
)

k
k 1

System Function and Frequency


Response
Frequency response of an LTI system
The DTFT of an impulse response is called the
frequency response of an LTI system, i.e.

H ( e j )

j n
h
(
n
)
e

H (e ) H (e ) e

j arg[ H ( e j )]

H (e j ) magnitude response function


arg[ H (e j )] phase response function

example
example

System Function and Frequency


Response

Group delays

d arg[ H (e j )]
g ( )
d

In general, the frequency response H (e j ) is a


complex function of

H (e j ) is a continuous function of

H (e j ) is a periodic function of , the period is 2

System Function and Frequency


Response

Response to exponential sequence

x ( n) e

j 0 n

H (e )

y( n) e

j 0 n

j 0

H (e

The output sequence is the input exponential sequence


modified by the response of the system at frequency

System Function and Frequency


Response

Response to sinusoidal sequences

x(n)

H (e )

y(n)

x ( n) A cos( 0 n )
y( n) A | H (e

j 0

) | cos( 0 n arg[ H (e

j 0

)])

x( n) Ak cos( k n k )
k

y( n) Ak | H (e j k ) | cos( k n k arg[ H (e j k )])


k

System Function and Frequency


Response

Response to arbitrary sequences

x(n)

H (e j ), h( n)

y( n) x ( n) h( n)
j

Y (e ) X (e ) H (e )

y(n)

System Function and Frequency


Response
Geometric interpretation of frequency response
M

Y (z)
H (z)

X (z)

m
b
z
m

m 0
N

k
a
z
k
k 0

Kz ( N M )

(z c
m 1
N

(z d
k 1

)
)

1
(
1

c
z
)

m
m 1
N

(1 d
k 1

z )

System Function and Frequency


Response
M

H (e ) Ke

j ( N M )

(e

cm )

(e

dk )

m 1
N
k 1

H (e ) e

j arg[ H ( e j )]

K is a real number

System Function and Frequency


Response
M

H (e ) K

(e

cm )

(e

dk )

m 1
N
k 1

zero vector
pole vector

Cm e

Dk e

cm m e
d k lk e

j m

j k

System Function and Frequency


Response
M

arg[ H (e )] arg[ K ] arg[e


j

m 1

arg[e

m 1

d k ] ( N M )

zero vector

C m e j c m m e j m

pole vector

Dk e j d k l k e j k

cm ]

System Function and Frequency


Response
M

H (e ) K

m 1
N

k 1

m 1

k 1

arg[ H (e )] arg[ K ] m k ( N M )
j

System Function and Frequency


Response
An approximate plot of the magnitude and phase
responses of the system function of an LTI system can
be developed by examining the pole and zero locations

To highly attenuate signal components in a specified


frequency range, we need to place zeros very close to
or on the unit circle in this range

To highly emphasize signal components in a


specified frequency range, we need to place poles very
close to or on the unit circle in this range

System Function and Frequency


m

Response
H (e j ) K m N1

k 1

1
2
2

l1

e j

1
1

l2

System Function and Frequency


Response
Minimum-Phase and Maximum-Phase system
M
N
H ( e j )
arg
m k ( N M )

K m 1
k 1

mi
mo
pi

the number of zeros inside the unit circle

po

the number of poles outside the unit circle

the number of zeros outside the unit circle


the number of poles inside the unit circle

M m i mo

N pi po

A causal stable system


H ( e j )
arg

po 0,

pi N

2m i 2pi 2 ( N M )
2

2 m i 2 M 2 mo
A causal stable system with all zeros inside the unit circle
is called a minimum-phase delayed system

H (e j )
arg

0
2

A causal stable system with all zeros outside the unit circle
is called a maximum-phase delayed system

H ( e j )
arg

2M
2

An anti-causal stable system

H ( e j )
arg

pi 0,

po N

2m i 2 ( N M )
2

An anti-causal stable system with all zeros inside the unit


circle is called a maximum-phase advanced system

H ( e j )
arg

2N 2po
2

An anti-causal stable system with all zeros outside the unit


circle is called a minimum-phase advanced system

H ( e j )
arg

2 ( N M ) 2 ( po mo )
2

System Function and Frequency


Response
Important properties of minimum-phase delayed system
minimum-phase delayed system is often called
minimum-phase system for short.
It plays an important role in telecommunications
Any nonminimum-phase system can be expressed as
the product of a minimum-phase system function and a
stable all-pass system

j
For all systems with the identical | H (e ) |

N 1

N 1

n 0

n 0

n0

n 0

2
2
|
h
(
n
)
|

|
h
(
n
)
|

min
2
2
|
h
(
n
)
|

|
h
(
n
)
|

min , m N-1

System Function and Frequency


Response
All-pass system

Definition

A system that has a constant magnitude response


for all frequencies, that is,

| H ap (e j ) | 1

The simplest example of an all-pass system is a


pure delay system with system function H ( z ) z k
This system passes all signals without modification
except for a delay of k samples.

System Function and Frequency


Response

1-th order all-pass system


1

z a
H ap ( z )
,
1
1 az
j

a re j , 0 r 1
j

j ( )

e re
1 re
j
H ap (e )
e

j j
j ( )
1 re e
1 re
j

1 r cos( ) jr sin( )

1
1 r cos( ) jr sin( )

System Function and Frequency


Response

An alternative form of 1-th order all-pass system

1 ( a ) 1 z 1
H ap ( z )
,
1
1 az
j

H ap (e ) r

a re j , 0 r 1
j Im[z ]
a

1
a
Re[z ]

Mirror image symmetry with


respect to the unit circle

System Function and Frequency


Response

2-th order all-pass system


1

z a
z a
j
H ap ( z )

, a re , 0 r 1
1
1
1 az 1 a z
H ap (e j ) 1
1 ( a ) 1 z 1 1 a 1 z 1
j
H ap ( z )

, a re , 0 r 1
1
1
1 az
1 a z
j

H ap (e ) r

example

System Function and Frequency


Response
example

let a re , 0 r 1

z 1 a z 1 a
z 2 2rz 1 cos r 2
H ap ( z )

1
1
1 az 1 a z
1 2rz 1 cos r 2 z 2
1
2 2
1 2rz cos r z
2
2 D( z )
z
z
1
2 2
D( z )
1 2rz cos r z
D( z ) 1 2rz cos r z
1

2 2

System Function and Frequency


Response

N-th order all-pass system


1

k
1

z a
H ap ( z )
k 1 1 a k z
N

( N 1 )

d N d N 1 z d 1 z
z
z D( z )

1
( N 1 )
N
1 d 1 z d N 1 z
dN z
D( z )
j

D( e ) D ( e
j

H (e ) 1

System Function and Frequency


Response

An alternative form for N-th order all-pass system

z ak
( z 1 bk )( z 1 bk )
H ap ( z )
1
1
1
k 1 1 a k z
k 1 (1 bk z )(1 bk z )
NR

NC

N R The number of real poles and zeros


N C The number of complex-conjugate pair of poles
and zeros

For causal and stable system

| a k | 1, | bk | 1

System Function and Frequency


Response

Application

Phase equalizers

When placed in cascade with a system that has


an undesired phase response, a phase equalizer
is designed to compensate for the poor phase
characteristics of the system and therefore to
produce an overall linear-phase response.

H ( z ) H ap ( z ) H d ( z )

System Function and Frequency


Response
j

H (e ) H ap (e ) H d (e )
j

H ap (e ) H d (e ) e

j [ ap ( ) d ( )]

arg[ H (e )] ap ( ) d ( )
Group delays

d arg[ H (e j )]
( )
ap ( ) d ( ) 0
d

System Function and Frequency


Response
Any causal-stable nonminimum-phase system can
be expressed as the product of a minimum-phase
delayed system cascaded with a stable all-pass system

example

H ( z ) H min ( z ) H ap ( z )
H ( z ) H 1 ( z ) ( z 1 zo ) ( z 1 zo )

H 1 ( z ) a minimum-phase system
1 1
, a pair of conjugate zeros outside the unit circle
zo zo
| zo | 1

1
1
1

z
z
1

z
z
o
o
H ( z ) H 1 ( z ) ( z 1 zo ) ( z 1 zo )

1
1
1 zo z 1 zo z
1

o
1

z zo z z
H 1 ( z ) (1 z z ) (1 zo z )

1
1 zo z 1 zo z
1
o

H min ( z ) H ap ( z )
H 1 ( z ) (1 zo z 1 ) (1 zo z 1 ) is a minimum-phase system
z 1 zo z 1 zo

1
1
1 zo z 1 zo z

is a 2-th all-pass system

System Function and Frequency


Response
By cascading an all-pass system an unstable
system can be made stable without changing its
magnitude response

example

z a
z a
H ap ( z )

1
1 az 1 a z 1

H ( z ) H ( z ) H ap ( z )
H (z ) unstable system
H (z ) stable system

System Function and Frequency


Response
Relationships between system representations

H (z )

Express H(z) in
z-1 cross multiply
and take inverse

Difference
Equation

ZT

take ZT solve
for Y/X

h(n )

substitute
j

Inverse DTFT

ze

Take DTFT
solve for Y/X

Inverse ZT

DTFT
j

H (e )

return

(n)

unit sample sequence

1
Amplitude

0.8
0.6
0.4
0.2
0
-10

-5

5
n

10

15

20

-5

5
n

10

15

20

( n 5)
1
Amplitude

0.8
0.6
0.4
0.2
0
-10

return

u(n)

unit step sequence


1

Amplitude

0.8
0.6
0.4
0.2
0
-5

u( n 5)

10

15

20

10

15

20

1
Amplitude

0.8
0.6
0.4
0.2
0
-5

return

R10 ( n)

Rectangular
sequence
unit step sequence

1
Amplitude

0.8
0.6
0.4
0.2
0
-5

R10 ( n 5)

10

15

20

10

15

20

1
Amplitude

0.8
0.6
0.4
0.2
0
-5

return

1.5 cos(0.05 2 n)
Amplitude

Sinusoidal sequence

1
0

-1
-2

10

20

30
n

40

50

60

10

20

30
n

40

50

60

1.5 sin( 0.05 2 n)


Amplitude

2
1
0

-1
-2

return

Real-valued
exponential
sequence
Sinusoidal
sequence

0.25

0.001 1.2

Amplitude

0.2

0.15
0.1

0.05
0

10

15
n

20

0.2

30

0.2 0.8 n

0.15

Amplitude

25

0.1

0.05
0

10

15
n

20

25

30

return

Complex-valued exponential sequence

real part

2e

Amplitude

1.5

1
( j )n
5 8

1
0.5
0
-0.5

10

20

25

30

35

25

30

35

imaginary part

1.5

Amplitude

15

1
0.5
0
-0.5

10

15

20

return

n 2 R1780( n 8)

finite-length sequence

Amplitude

60
40
20
0
-20

-15

-10

cos(0.4n)

10

15

20 n

10

15

20 n

infinite-length sequence

Amplitude

-5

0.5
0

-0.5
-1
-20

-15

-10

-5

return

right-sided sequence

0.8

0.2 0.8 n u( n 5)

Amplitude

0.6
0.4
0.2
0
-10

-5

10
n

15

20

causal sequence

0.2

30

0.2 0.8 n u( n)

0.15

Amplitude

25

0.1

0.05
0
-10

-5

10
n

15

20

25

30

return

left-sided sequence

0.8

0.2 0.8 n u( n 5)

Amplitude

0.6
0.4
0.2
0
-30

-25

-20

-15

-10
n

-5

10

0.2 0.8 n u( n)

anti-causal sequence

0.2

Amplitude

0.15
0.1

0.05
0
-30

-25

-20

-15

-10
n

-5

10

return

0.2 0.8

two-sided sequence

0.2

|n|

Amplitude

0.15
0.1

0.05
0
-20

-15

-10

0
n

10

15

20

0.2 cos(0.3n)

two-sided sequence

0.2

Amplitude

-5

0.1
0

-0.1
-0.2
-20

-15

-10

-5

0
n

10

15

20

return

absolutely summable sequence

0.3 n u( n)

Amplitude

0.8
0.6
0.4
0.2
0
-5

10

15
n

20

25

30

35

absolutely summable sequence

0.85 n u( n)

Amplitude

0.8
0.6
0.4
0.2
0
-5

10

15
n

20

25

30

35

return

0.15

Amplitude

sin 0.3n
n

square-summable sequence

0.1
0.05
0

-0.05
-20

-15

-10

-5

0
n

10

15

square-summable sequence

0.2

sin 0.6n
n

0.15
Amplitude

20

0.1
0.05
0

-0.05
-20

-15

-10

-5

0
n

10

15

20

return

Time-shifting operation

0.2 0.8 u( n)
n

original sequence

Amplitude

0.2

0.1

0
-10

-5

Amplitude

0.2

10
15
delayed sequence

20

25

30

0.2 0.8 n 5 u( n 5)

0.1

0
-10

-5

Amplitude

0.2

5
10
15
advanced sequence

20

25

30 n

0.2 0.8 n 5 u( n 5)

0.1

0
-10

-5

10

15

20

25

30

return

folding operation

original sequence

0.8 n u( n)

Amplitude

0.8
0.6
0.4
0.2
0
-20

-15

-10

-5

0
n

10

15

20

folding sequence

0.8 u( n)

Amplitude

0.8
0.6
0.4
0.2
0
-20

-15

-10

-5

0
n

10

15

20

return

addition operation

x1(n)

1
Amplitude

0.8 n u( n)

0.5

10

15

Amplitude

25

30

35

40 n

cos(0.2n)u( n)

-1

10

15

20
x1(n)+x2(n)

25

30

35

40 n

10

15

20

25

30

35

40 n

2
Amplitude

20
x2(n)

0.8 n u( n) cos(0.2n)u( n)

1
0

-1

return

modulation operation
Amplitude

0.1

-0.1

20

40

60

80
x2(n)

100

120

20

40

60

80
x1(n)*x2(n)

100

120

20

40

60

80

100

120

Amplitude

140

160

140

160

140

160

sin 0.125n

-1
0.1
Amplitude

0.1 sin 0.0125n

x1(n)

x1 ( n) x 2 ( n)

-0.1

return

periodic sequence
1

Amplitude

sin( n)
8

periodic sequence

0.5
0

-0.5
-1

10

20

30

50

60

70

80

90

sin( n)
16

periodic sequence

Amplitude

40

0.5
0

-0.5
-1

10

20

30

40

50

60

70

80

90

return

Periodicity of sequence

x1(n)

Amplitude

10

20

30

40

10

20

30

40

1
Amplitude

n)
8

-1

x2(n)

50

60

70

80

50

60

70

80

90

sin(

3
n)
10

-1
1
Amplitude

sin(

x3(n)

90

sin( 0.4n)

-1

10

20

30

40

50

60

70

80

90

return

T0 32

T
3

Periodicity of sequence 2

Amplitude

0.5
0

-0.5
-1

10

20

30

40

50

T0 10

T
3

Amplitude

60

0.5
0

-0.5
-1

10

20

30

40

50

60

return

12

( n) 4 ( n 1)
3 ( n 2) 6 ( n 3)
8 ( n 4) 10 ( n 5)
11 ( n 6) 9 ( n 7 )

11
10
9
8
7
6
5
4
3
2
1
0

10 11 12 13 14 15 16 17 18 19 20

return

x(n)

Amplitude

10

0 .9

0
-20

-15

-10

-5

0
xe(n)

10

15

20

-15

-10

-5

0
xo(n)

10

15

20

-15

-10

-5

10

15

20

Amplitude

6
4
2
0
-20

Amplitude

-5
-20

return

Accumulator
n

n 1

y( n)

x(l ) x(l ) x(n) y(n 1) x(n)

The output y(n) at time instant n is the sum of the input


sample x (n) at time instant n and the previous output y( n 1)
at time instant n-1, which is the sum of all previous input
sample values from to n-1
The input-output relation can also be written in the form:

y ( n)

l 0

l 0

x(l ) x(l ) y( 1) x(l ), n 0

This form is used for a causal input sequence, in which


case y(-1) is called the initial condition
return

M-point moving-average system

y( n)

1
M

M 1

x( n k )
k 0

An application: consider x ( n) s( n) d ( n)
Where s(n) is the signal, and d (n) is a random noise

s( n) 2n0.9 n

M 8

1 7
y( n) x ( n k )
8 k 0

return

Amplitude

10

10

15

20

25
x(n)

30

35

40

45

50

10

15

20

25
y(n)

30

35

40

45

50

10

15

20

25

30

35

40

45

50

Amplitude

10

0
10
Amplitude

s( n) 2n0.9 n

s(n),d(n)

return

y ( n)

Accumulator

x( l )

T [ x1 ( n)]

x (l ),

T [ax1 ( n) bx 2 ( n)]
n

T [ x 2 ( n)]

x (l )

[ax ( l ) bx

( l )]

a x1 ( l ) b x 2 ( l ) aT [ x1 ( n)] bT [ x 2 ( n)]
Hence, the above system is linear

return

y ( n)

Accumulator

x( l )

n k

T [ x ( n k )]

x(l )

y( n k )

n k

x(l ) T [ x(n k )]

Hence, the above system is time-invariant

return

1.5
Amplitude

x ( n) R10 ( n)

x(n)

0.5
0

10

15

20

Amplitude

1.5

25
h(n)

30

35

40

45

50

h( n) 0.9 n u( n)

0.5
0

10

15

20

Amplitude

10

25
y(n)

30

35

40

45

50

y( n) x ( n) h( n)

10

15

20

25

30

35

40

45

50

return

The M-point moving average filter


1 M 1
y( n)
x(n k )

M k 0
For a bounded input x ( n) B x , we have

1
y( n)
M

M 1

1
x
(
n

k
)

M
k 0

M 1

x( n k )
k 0

( MB x ) B x
M
Hence, the M-point moving average filter is BIBO stable

return

A causal LTI discrete-time system

h( n) u( n)
n

1
S | a | u( n) | a |
1 a
n
n0
n

if | a | 1

if | | 1, the system is stable


if | | 1, the system is not stable

return

DTFT[0.9 n e jn / 3 R11 ( n)]


real part

Amplitude part

8
6

Amplitude

Amplitude

6
4

2
0
-2

-1

Amplitude
-1

0
pi

-1

0
pi

imaginary part

-0.5
-2

0
-2

phase part

0.5

phase(pi)

0
pi

-5
-2

-1

0
pi

return

Amplitude part

DTFT[0.9 n R11 ( n)]


8
6

Amplitude

Amplitude

6
4

2
0
-1

-0.5

0.5

0
-1

Amplitude

phase(pi)

0
pi
phase part

0.5

-0.5
-1

real part

-0.5

0
pi

0.5

-0.5

0
pi

0.5

imaginary part

-5
-1

-0.5

0
pi

0.5

return

H (z)
1

z 1
z 2 0.9 z 0.81

0.8
0.6

Imaginary Part

0.4
0.2
0

-0.2
-0.4
-0.6
-0.8
-1
-1

-0.5

0
Real Part

0.5

Amplitude response

15

H (z)

z 1
z 2 0.9 z 0.81

Amplitude

10

0.1

0.2

0.3

0.4

0.5
pi

0.6

0.7

0.8

0.9

0.7

0.8

0.9

phase response

phase(pi)

0.5
0

-0.5
-1

0.1

0.2

0.3

0.4

0.5
pi

0.6

return

7 19 1

z
H ( z ) 12 24
5 1
1 z z 2
2

Imaginary Part

0.5

-0.5

-1

-1

-0.5

0.5
Real Part

1.5

Amplitude response

0.55

Amplitude

0.5
0.45

7 19 1

z
H ( z ) 12 24
5 1
1 z z 2
2

0.4

0.35
0.3

0.1

0.2

0.3

0.4

0.5
pi

0.6

0.7

0.8

0.9

0.7

0.8

0.9

phase response

phase(pi)

-0.1
-0.2
-0.3
-0.4

0.1

0.2

0.3

0.4

0.5
pi

0.6

return

Imaginary Part

z a
z a
H ap ( z )

1
1 az 1 a z 1
1.5
1

(r , )
21
3

1
a

0.5

-0.5

Mirror -1image symmetry

1
a

-1.5
-2

-1.5

-1

-0.5

0
Real Part

0.5

1.5

return

1 2.61z 1 3.08 z 2 1.85 z 3 0.83 z 4


H (z)
1
2
1

1
.
12
z

0
.
48
z
1
0.8

1
zo

0.6

Imaginary Part

0.4
0.2
2

-0.2
-0.4

1
zo

-0.6
-0.8
-1
-1

-0.5

0
Real Part

0.5

return

1
0.8

1
zo

0.6

Imaginary Part

0.4

zo

0.2
2

-0.2

zo

-0.4

1
zo

-0.6
-0.8
-1
-1

-0.5

0
Real Part

0.5

return

1 0.83 z 1 0.35 z 2
H (z)
1
2
1

2
.
41
z

2
.
26
z
1

a
1

0.8
0.6

zo

Imaginary Part

0.4

0.2
0

-0.2
-0.4
-0.6

1
a

-0.8
-1
-1

-0.5

0
Real Part

0.5

return