DSP Lec 1

Lecture 1: Introduction and Review
EE 4541 Digital Signal Processing, Fall 2011

EE 4541 Digital Signal Processing
Lecture 1: Introduction and Review of Discrete-time
signals and systems

Emad S. Ebbini
Department of Electrical and Computer Engineering
2
Introduction
DSP is concerned with the analysis and
synthesis of discrete-time signals (sequences)
in digital form.
This includes the analysis and design of
systems and operators for:
Enhancement (e.g., improve SNR)
Correction of Distortions (e.g., equalization of
communication channels)
Parameter Estimation (e.g., Doppler shift)
DSP vs SP?

3
Introduction
For signals with bandwidths in the audio range, DSP
offers a superior alternative to analog signal processing.
HDTV provides an excellent example of DSP applications
in high BW video applications.
Realtime implementation of DSP algorithms is enabled
by programmable digital signal processors (DSPs).
A variety of options: GPDSPs, FPGAs, GPGPUs, ASICs,
etc. (EE5545, Spring13)
Some extremely high BW applications still require optical
components and/or surface acoustic wave devices for
(analog) processing.
4
Digital Processing of Analog Signals
Advantages:
Flexibility; Robustness; Self test; Controlled Error
5
Need to account for:
Sampling
Quantization
Analog Antialiasing Filter
Analog-to-Digital Conversion (A/D)
=
=
n
nT t t s ) ( ) ( o
Antialiasing
Filter
) (t x
a
] [ ) ( ] [ n e nT x n x
q a
+ =
Sampling Function
Oversampling and Noise Shaping
6
Digital Signal Processing
Filtering:
Removing or reducing noise.
Enhancement of specific frequency bands (equalization).
Multirate DSP (Upsampling & Downsampling)
Inverse filtering or deconvolution.
Frequency Domain Methods:
Spectral estimation using DFT.
Direction of arrival (Array Processing).
Range/Doppler Radar (Time-Frequency Analysis).
Tracking Methods
Cross-correlation and time delay estimation
Motion tracking, e.g. for video compression (spatio-temporal)
Multi-dimensional DSP, e.g. image and video processing.

7
Digital Filter Structure
| | | | | |

= =
+ =
N
k
k
N
k
k
k n x b k n y a n y
0 1
Direct Form II
Implementation
To change the
filter, we simply
need to change
the coefficients.
But how to
determine these
coefficients?
Filter Design
MATLAB Implementation: y=filter(B,A,x);
[y,zf]=filter(B,A,x,zi);
8
Digital-to-Analog Conversion
Need to account for:
Non-ideal reconstruction (interpolation) filter
Oversampling and Noise Shaping
9
Reading Assignment (Week 1)
Chapter 1: Survey. I will plan on
discussing questions related to this
chapter on Thursday.
Chapter 2: Sections 2.1 2.5. Review of
prerequisite material.
Section 2.10 will be covered briefly.
10
Discrete-Time Signals:
Time-Domain Representation
Signals represented as a sequence of
numbers, called samples
Sample value o a typical signal or
sequence denoted as []with being an
integer, < < .
[]defined only for integer values of
and undefined for noninteger values of
Discrete-time signal represented by []
11
DT signal may also be written as a
sequence of numbers
[] = , 0.2,2.2,1.1,0.2, 3.7,2.9,

Here, 1 = 0.2, 0 = 2.2, 1 ,etc.
Samples may arise from A/D conversion
=
< <
is the sampling period (uniform, ideal
sampling)
12
Graphical representation of a discrete-time
signal with real-valued samples:

MATLAB implementation: stem(x),
stem(n,x), stem(n,x,filled)
13

14
Here, the nth sample is given by
=

=
=
, < <
The spacing, T, between samples is called
the sampling interval or sampling period
Reciprocal of the sampling interval
denoted as
, is called the sampling

frequency
=
1

15
Unit of sampling frequency is cycles per
second, or hertz (Hz), if T is in seconds
Whether or not the sequence {x[n]} has been
obtained by sampling, the quantity x[n] is
called the nth sample of the sequence
{x[n]} is a real sequence, if the nth sample
x[n] is real for all values of n
Otherwise, {x[n]} is a complex sequence
16
A complex sequence {x[n]} can be written as
where
and are the real and imaginary
parts of x[n]
The complex conjugate sequence of {x[n]} is
given by
Often the braces are ignored to denote a
sequence if there is no ambiguity
] [n x
re
] [n x
im
]} [ { ]} [ { ]} [ { n x j n x n x
im re
+ =
]} [ { ]} [ { ]} [ * { n x j n x n x
im re
=
17
Two types of discrete-time signals:
- Sampled-data signals in which samples
are continuous-valued (infinite precision)
- Digital signals in which samples are
discrete-valued (quantized)
Signals in a practical digital signal processing
system are digital signals obtained by
quantizing the sample values either by
rounding or truncation
18
A discrete-time signal may be a finite-
length or an infinite-length sequence
Finite-length (also called finite-duration
or finite-extent) sequence is defined
only for a finite time interval:
where and with
Length or duration of the above finite-
length sequence is
2 1
N n N s s
1
N < <
2
N
2 1
N N s
1
1 2
+ = N N N
19
A length-N sequence is often referred to
as an N-point sequence
The length of a finite-length sequence
can be increased by zero-padding,
i.e., by appending it with zeros
A right-sided sequence x[n] has
zero-valued samples for
If a right-sided sequence is
called a causal sequence

1
N n <
, 0
1
> N
20
A left-sided sequence x[n] has zero-
valued samples for
If a left-sided sequence is
called a anti-causal sequence
A right-sided sequence

A left-sided sequence
2
N n >
, 0
2
s N
21
Operations on Sequences:
Basic Operations
Product (modulation) operation:

Modulator

An application is in forming a finite-length sequence
from an infinite-length sequence by multiplying the
latter with a finite-length sequence called an
window sequence
Process called windowing
x[n]
y[n]
w[n]
] [ ] [ ] [ n w n x n y =
22
Basic Operations
Addition operation:

Adder

Multiplication operation

Multiplier
] [ ] [ ] [ n w n x n y + =
A
x[n]
y[n]
] [ ] [ n x A n y =
x[n]
y[n]
w[n]
+
23
Basic Operations
Time-shifting operation:
where N is an integer
If N > 0, it is delaying operation
Unit delay

If N < 0, it is an advance operation
Unit advance
] [ ] [ N n x n y =
y[n]
x[n] z
1
z
y[n]
x[n]
] [ ] [ 1 = n x n y
] [ ] [ 1 + = n x n y
24
Basic Operations
Time-reversal (folding) operation:

Branching operation: Used to provide
multiple copies of a sequence
] [ ] [ n x n y =
x[n]
x[n]
x[n]
25
Combinations of Basic Operations

Example Weighted Moving Average
] 3 [ ] 2 [ ] 1 [ ] [ ] [
4 3 2 1
+ + + = n x n x n x n x n y o o o o
26
Sampling Rate Alteration:
Basic Operations
Employed to generate a new sequence y[n]
with a sampling rate higher or lower
than that of the sampling rate of a given
sequence x[n]
Sampling rate alteration ratio is

If R > 1, the process called interpolation
If R < 1, the process called decimation

s
F
'
s
F
s
s
F
F
R
'
=
27
Basic Operations
In up-sampling by an integer factor L > 1,
equidistant zero-valued samples are
inserted by the up-sampler between each
two consecutive samples of the input
sequence x[n]:
1 L
=
=
otherwise , 0
, 2 , , 0 ], / [
] [
L L n L n x
n x
u
L ] [n x
] [n x
u
28
Basic Operations
An example of the up-sampling operation (L =3)

0 10 20 30 40 50
-1
-0.5
0
0.5
1
Output sequence up-sampled by 3
Time index n
A
m
p
l
i
t
u
d
e
0 10 20 30 40 50
-1
-0.5
0
0.5
1
Input Sequence
Time index n
A
m
p
l
i
t
u
d
e
29
Basic Operations
In down-sampling by an integer factor
M > 1, every M-th sample of the input
sequence is kept and in-between
samples are removed:
1 M
] [ ] [ nM x n y =
] [n x
] [n y
M
30
Basic Operations
An example of the down-sampling operation (M =3)
0 10 20 30 40 50
-1
-0.5
0
0.5
1
Output sequence down-sampled by 3
A
m
p
l
i
t
u
d
e
Time index n
0 10 20 30 40 50
-1
-0.5
0
0.5
1
Input Sequence
Time index n
A
m
p
l
i
t
u
d
e
31
Classification of Sequences Based on
Symmetry

Conjugate-symmetric sequence:

If x[n] is real, then it is an even
sequence

] [ * ] [ n x n x =
32
Symmetry
Conjugate-antisymmetric sequence:

If x[n] is real, then it is an odd sequence
] [ * ] [ n x n x =
35
Periodicity
A sequence satisfying
is called a periodic sequence with a period N
where N is a positive integer and k is any
integer.
Smallest value of N satisfying
is called the fundamental period

Otherwise, it is called an aperiodic sequence
] [
~
n x ], [
~
] [
~
kN n x n x + =
= ] [
~
n x ] [
~
kN n x +
36
Classification of Sequences:
Energy and Power Signals
Total energy of a sequence x[n] is defined by

An infinite length sequence with finite sample
values may or may not have finite energy
The average power of an aperiodic sequence
is defined by
=

= n
n x
2
x
] [ c
=
=
+

K
K n
K
K
n x P
2
1 2
1
x
] [ lim
37
Define the energy of a sequence x[n] over a
finite interval as
Then

The average power of a periodic sequence
with a period N is given by

The average power of an infinite-length
sequence may be finite or infinite
=
=
K
K n
K x
n x
2
,
] [ c K n K s s
K x
K
K
x
P
,
1 2
1
lim c
+

=
=

=
1
0
2
1
] [
N
n
N
x
n x P
38
Example - Consider the causal sequence
defined by

Note: x[n] has infinite energy
Its average power is given by
5 4
1 2
1 9
1 9
1 2
1
0
.
) (
lim lim =
+
+
=
|
|
.
|
\
|
+
=

=

K
K
K
P
K
K
n
K
x
<
>
=
0 0
0 1 3
n
n
n x
n
,
, ) (
] [
39
An infinite energy signal with finite average
power is called a power signal
Example - A periodic sequence which has a finite
average power but infinite energy
A finite energy signal with zero average power is
called an energy signal
Example - a finite-length sequence which has finite
energy but zero average power
40
Other Types of Classifications
A sequence x[n] is said to be bounded if

A sequence x[n] is said to be absolutely
summable if

A sequence x[n] is said to be square-
summable if
< s
x
B n x ] [
<
= n
n x ] [
<
= n
n x
2
] [
41
Unit sample sequence -

Unit step sequence -
Basic Sequences
=
=
=
0 , 0
0 , 1
] [
n
n
n o
<
>
=
0 , 0
0 , 1
] [
n
n
n u
42
Basic Sequences
Real sinusoidal sequence -

where A is the amplitude, is the
angular frequency, and is the phase
of x[n]
Example -
) cos( ] [ | e + = n A n x
o
o
e
|
0 10 20 30 40
-2
-1
0
1
2
Time index n
A
m
p
l
i
t
u
d
e
e
0
= 0.1t
43
Basic Sequences
Complex exponential sequence -

where A and are real or complex
numbers
If we write
then we can express

where
, ] [
n
A n x o =
< < n
o
,
) (
o
j
o
e
e o
o
+
=
,
| j
e A A=
], [ ] [ ] [
) (
n x j n x e e A n x
im re
n
o
j
o
j
+ = =
+ e o |
), cos( ] [ | e
o
+ = n e A n x
o
n
o
j
re
) sin( ] [ | e
o
+ = n e A n x
o
n
o
im
44
Basic Sequences
and of a complex
exponential sequence are real
sinusoidal sequences with constant
, growing , and decaying
amplitudes for n > 0. E.g.
] [n x
re
] [n x
im
( ) 0 =
o
o
( ) 0 >
o
o
( ) 0 <
o
o
n j n x ) exp( ] [
6 12
1 t
+ =
0 10 20 30 40
-1
-0.5
0
0.5
1
Time index n
A
m
p
l
i
t
u
d
e
Real part
0 10 20 30 40
-1
-0.5
0
0.5
1
Time index n
A
m
p
l
i
t
u
d
e
Imaginary part
45
Basic Sequences
Real exponential sequence -

where A and o are real numbers
, ] [
n
A n x o = < < n
0 5 10 15 20 25 30
0
10
20
30
40
50
Time index n
A
m
p
l
i
t
u
d
e
o = 1.2
0 5 10 15 20 25 30
0
5
10
15
20
Time index n
A
m
p
l
i
t
u
d
e
o = 0.9
46
Basic Sequences
Sinusoidal sequence and complex
exponential sequence
are periodic sequences of period N if
where N and r are positive integers
Smallest value of N satisfying
is the fundamental period of the sequence
To verify the above fact, consider

) cos( | e + n A
o
) exp( n j B
o
e
r N
o
t e 2 =
r N
o
t e 2 =
) cos( ] [ | e + = n n x
o 1
) ) ( cos( ] [ | e + + = N n n x
o 2
47
Basic Sequences
Now

which will be equal to
only if
and
These two conditions are met if and only if
or
) ) ( cos( ] [ | e + + = N n n x
o 2
N n N n
o o o o
e | e e | e sin ) sin( cos ) cos( + + =
] [ ) cos( n x n
o 1
= +| e
0 = N
o
e sin
1 = N
o
e cos
r N
o
t e 2 =
r
N
o
=
e
t 2
48
Basic Sequences
If is a noninteger rational
number, then the period will be a multiple
of
Otherwise, the sequence is aperiodic
Example - is an
aperiodic sequence
o
e t / 2
o
e t / 2
) sin( ] [ | + = n n x 3
49
Basic Sequences

Here

Hence (for r = 0)

0 =
o
e
1
0
2
= =
r
N
t
0 10 20 30 40
0
0.5
1
1.5
2
Time index n
A
m
p
l
i
t
u
d
e
e
0
= 0
50
Basic Sequences

Here

Hence for r = 1
t e 1 . 0 =
o
20
1 . 0
2
= =
t
tr
N
0 10 20 30 40
-2
-1
0
1
2
Time index n
A
m
p
l
i
t
u
d
e
e
0
= 0.1t
51
Basic Sequences
Property 1 - Consider
and
with and
where k is any
positive integer
If then x[n] = y[n]
Thus, x[n] and y[n] are indistinguishable
) exp( ] [
1
n j n x e =
) exp( ] [
2
n j n y e =
t e < s
1
0
) 1 ( 2 2
2
+ < s k k t e t
, 2
1 2
k t e e + =
52
Basic Sequences
Property 2 - The frequency of oscillation of
increases as increases from 0
to t, and then decreases as increases from
to
Thus, frequencies in the neighborhood of
are called low frequencies, whereas,
frequencies in the neighborhood of are
called high frequencies
) cos( n A
o
e
o
e
o
e
t
t 2
t e =
0 = e
53
Basic Sequences
Because of Property 1, a frequency in the
neighborhood of e = 2t k is indistinguishable
from a frequency in the neighborhood
of e = 0
and a frequency in the neighborhood of
is indistinguishable from a
frequency in the neighborhood
of e = t
o
e
k
o
t e 2
o
e
) ( 1 2 + k
o
t e
) ( 1 2 + = k t e
54
Basic Sequences
Frequencies in the neighborhood of e = 2t k
are usually called low frequencies
Frequencies in the neighborhood of e = t
(2k+1) are usually called high frequencies
is a low-
frequency signal
is a high-
frequency signal

) . cos( ) . cos( ] [ n n n v t t 9 1 1 0
1
= =
) . cos( ) . cos( ] [ n n n v t t 2 1 8 0
2
= =
55
An arbitrary sequence can be represented in
the time-domain as a weighted sum of some
basic sequence and its delayed (advanced)
versions

Basic Sequences
] 2 [ ] 1 [ 5 . 1 ] 2 [ 5 . 0 ] [ + + = n n n n x o o o
] 6 [ 75 . 0 ] 4 [ + + n n o o
Biomedical Signals
56
0 200 400 600 800 1000 1200 1400 1600 1800 2000
0
500
1000
A
/
D

C
o
u
n
t
ECG Waveform
0 200 400 600 800 1000 1200 1400 1600 1800 2000
0
100
200
m
m
H
g
Aortic Pressure
0 200 400 600 800 1000 1200 1400 1600 1800 2000
0
100
200
m
m
H
g
Radial Pressure
10 seconds
of data.
What is F
s
?

DSP Lec 1

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

DSP Lec 1

Hochgeladen von

Copyright:

Verfügbare Formate

Lecture 1: Introduction and Review

EE 4541 Digital Signal Processing, Fall 2011

Lecture 1: Introduction and Review

, is called the sampling

Das könnte Ihnen auch gefallen