Time-Domain Characterization of LTI Discrete-Time Systems
Time-Domain Characterization of LTI Discrete-Time Systems
Time-Domain Representation
What is a signal ?
• Biomedicine
EEG, ECG, MRI, X-Rays, Ultrasounds
• Seismology
tectonic plate movement, earthquake prediction
• Economics
stock market data
What is DSP?
Mathematical and algorithmic manipulation of discretized and
quantized or naturally digital signals in order to extract the most
relevant and pertinent information that is carried by the signal.
What is a signal?
What is a system?
What is processing?
Signals can be characterized in several ways
Continuous time signals vs. discrete time signals (x(t), x[n]).
Temperature in London / signal on a CD-ROM.
Continuous valued signals vs. discrete signals.
Amount of current drawn by a device / average scores of TOEFL in a
school over years.
–Continuous time and continuous valued : Analog signal.
–Continuous time and discrete valued: Quantized signal.
–Discrete time and continuous valued: Sampled signal.
–Discrete time and discrete values: Digital signal.
Real valued signals vs. complex valued signals.
Resident use electric power / industrial use reactive power.
Scalar signals vs. vector valued (multi-channel) signals.
Blood pressure signal / 128 channel EEG.
Deterministic vs. random signal:
Recorded audio / noise.
One-dimensional vs. two dimensional vs. multidimensional
signals.
Speech / still image / video.
Systems
• For our purposes, a DSP system is one that can mathematically
manipulate (e.g., change, record, transmit, transform) digital
signals.
• Furthermore, we are not interested in processing analog signals
either, even though most signals in nature are analog signals.
Various Types of Processing
Modulation and demodulation.
Signal security.
Encryption and decryption.
Multiplexing and de-multiplexing.
Data compression.
Signal de-noising.
Filtering for noise reduction.
Speaker/system identification.
Signal enhancement –equalization.
Audio processing.
Image processing –image de-noising, enhancement,
watermarking.
Reconstruction.
Data analysis and feature extraction.
Frequency/spectral analysis.
Filtering
• By far the most commonly used DSP operation
Filtering refers to deliberately changing the frequency content of the signal,
typically, by removing certain frequencies from the signals.
For de-noising applications, the (frequency) filter removes those frequencies
in the signal that correspond to noise.
In various applications, filtering is used to focus to that part of the spectrum
that is of interest, that is, the part that carries the information.
256x256 64x64
Discrete (Sampled) and Digital (Quantized) Image
Discrete (Sampled) and Digital (Quantized) Image
Discrete (Sampled) and Digital (Quantized) Image
n
N1
A right-sided sequence
• If a right-sided sequence is called a
N1 0,
causal sequence
Discrete-Time Signals:
Time-Domain Representation
• A left-sided sequence x[n] has zero-valued
samples for n N 2
N2
n
A left-sided sequence
• Multiplication operation
A
– Multiplier x[n] y[n] y[n] A x[n]
Basic Operations
• Time-shifting operation: y[n] x[n N ]
where N is an integer
• If N > 0, it is a delay operation
– Unit delay y[n] x[n 1]
x[n] z 1 y[n]
x[n]
Combinations of Basic
Operations
• Example -
x[n] L xu [n]
Sampling Rate Alteration
• An example of the up-sampling operation
Input Sequence Output sequence up-sampled by 3
1 1
0.5 0.5
Amplitude
Amplitude
0 0
-0.5 -0.5
-1 -1
0 10 20 30 40 50 0 10 20 30 40 50
Time index n Time index n
Sampling Rate Alteration
• In down-sampling by an integer factor
M > 1, every M-th samples of the input
sequence are kept and M 1 in-between
samples are removed:
y[n] x[nM ]
x[n] M y[n]
Sampling Rate Alteration
• An example of the down-sampling
operation
Input Sequence Output sequence down-sampled by 3
1 1
0.5 0.5
Amplitude
0 Amplitude 0
-0.5 -0.5
-1 -1
0 10 20 30 40 50 0 10 20 30 40 50
Time index n Time index n
Classification of Sequences
Based on Symmetry
• Conjugate-symmetric sequence:
x[n] x * [ n]
If x[n] is real, then it is an even sequence
An even sequence
Classification of Sequences
Based on Symmetry
• Conjugate-antisymmetric sequence:
x[n] x * [ n]
If x[n] is real, then it is an odd sequence
An odd sequence
Classification of Sequences
Based on Symmetry
• It follows from the definition that for a
conjugate-symmetric sequence {x[n]}, x[0]
must be a real number
• Likewise, it follows from the definition that
for a conjugate anti-symmetric sequence
{y[n]}, y[0] must be an imaginary number
• From the above, it also follows that for an
odd sequence {w[n]}, w[0] = 0
Classification of Sequences
Based on Symmetry
• Any complex sequence can be expressed as a
sum of its conjugate-symmetric part and its
conjugate-antisymmetric part:
x[n] xcs [n] xca [n]
where
xcs [n] 12 x[n] x * [ n]
n
–4 –3 –2 –1 0 1 2 3 4 5 6
n
–4 –3 –2 –1 0 1 2 3 4 5 6
Basic Sequences
• Real sinusoidal sequence -
x[n] A cos(o n )
where A is the amplitude, o is the angular
frequency, and is the phase of x[n]
Example - w = 0.1
o
2
1
Amplitude
-1
-2
0 10 20 30 40
Time index n
Basic Sequences
• Complex exponential sequence -
n
x[n] A , n
where A and are real or complex numbers
( o jo ) j
• If we write e , A A e ,
then we can express
j ( o jo ) n
x[n] A e e xre [n] j xim [n],
where
o n
xre [n] A e cos(o n ),
o n
xim [n] A e sin(o n )
Basic Sequences
• xre [n] and xim [n] of a complex exponential
sequence are real sinusoidal sequences with
constant o 0 , growing o 0 , and
decaying o 0 amplitudes for n > 0
Real part Imaginary part
1 1
0.5 0.5
Amplitude
Amplitude
0 0
-0.5 -0.5
-1 -1
0 10 20 30 40 0 10 20 30 40
Time index n Time index n
1
x[n] exp( j 6 )n 12
Basic Sequences
• Real exponential sequence -
x[n] A n , n
where A and are real or complex numbers
a = 1.2 a = 0.9
50 20
40
15
Amplitude
Amplitude
30
10
20
10 5
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time index n Time index n
Basic Sequences
• Sinusoidal sequence A cos(o n ) and
complex exponential sequence B exp( jo n)
are periodic sequences of period N if o N 2r
where N and r are positive integers
• Smallest value of N satisfying o N 2r
is the fundamental period of the sequence
• To verify the above fact, consider
x1[n] cos(o n )
x2 [n] cos(o (n N ) )
Basic Sequences
• Now x2 [n] cos(o (n N ) )
cos(o n ) cos o N sin(o n ) sin o N
which will be equal to cos(o n ) x1[n]
only if
sin o N 0 and cos o N 1
• These two conditions are met if and only if
o N 2 r 2 N
or r
o
Basic Sequences
1.5
Amplitude
1
0.5
0
• Here 0 10 20
Time index n
30 40
o 0
• Hence period for r = 0
2 r
N 1
0
Basic Sequences
w = 0.1p
0
2
Amplitude 0
-1
-2
0 10 20 30 40
Time index n
• Here o 0.1
2 r
• Hence N 20 for r = 1
0.1
Basic Sequences
• Property 1 - Consider x[n] exp( j1n)and
y[n] exp( j2 n) with 0 1 and
2k 2 2(k 1) where k is any positive
integer
• If 2 1 2k ,then x[n] = y[n]
0.5
Amplitude
-0.5
-1
0 0.2 0.4 0.6 0.8 1
• Note that each sequence has exactly the same
time
-2
0 10 20 30 40 50
Time index n
7
s[n]
6 y[n]
5
Amplitude
0
0 10 20 30 40 50
Time index n
Discrete-Time Systems:Examples
• Linear interpolation - Employed to estimate
sample values between pairs of adjacent
sample values of a discrete-time sequence
• Factor-of-4 interpolation
y[n]
3 4
n
0 1 2 5 6 7 8 9 10 11 12
Discrete-Time Systems:
Examples
• Factor-of-2 interpolator -
• Consider
2
y[n] x [n] x[n 1]x[n 1]
• Outputs y1[n] and y2 [n] for inputs x1[n]
and x2 [n] are given by
y1[n] x12 [n] x1[n 1]x1[n 1]
] impulse
h•[nThe 1 [n] 2 [n 1is
response 3 a[nfinite-length
] thus 2] 4 [n 3]
sequence of length 4 given by
{h[n]} {1, 2 , 3 , 4}
Impulse Response
• Example - The impulse response of the
discrete-time accumulator
n
y[n] x[]
is obtained by setting x[n] = [n] resulting
in n
h[n] [] [n]
Impulse Response
• Example - The impulse response {h[n]} of
the factor-of-2 interpolator
1
y[n] xu [n] ( xu [n 1] xu [n 1])
2
is obtained by setting xu [n] [n] and is
given by
1
h[n] [n] ( [n 1] [n 1])
2
• The impulse response is thus a finite-length
sequence of length 3:
{h[n]} {0.5, 1 0.5}
Time-Domain Characterization
of LTI Discrete-Time System
• Input-Output Relationship -
It can be shown that a consequence of the
linear, time-invariance property is that an
LTI discrete-time system is completely
characterized by its impulse response
• Knowing the impulse response one
can compute the output of the system for
any arbitrary input
Time-Domain Characterization
of LTI Discrete-Time System
• Let h[n] denote the impulse response of a
LTI discrete-time system
• We compute its output y[n] for the input:
x[n] 0.5[n 2] 1.5[n 1] [n 2] 0.75[n 5]
• As the system is linear, we can compute its
outputs for each member of the input
separately and add the individual outputs to
determine y[n]
Time-Domain Characterization
of LTI Discrete-Time System
• Since the system is time-invariant
input output
[n 2] h[n 2]
[n 1] h[n 1]
[n 2] h[n 2]
[n 5] h[n 5]
Time-Domain Characterization
of LTI Discrete-Time System
• Likewise, as the system is linear
input output
0.5[n 2] 0.5h[n 2]
1.5[n 1] 1.5h[n 1]
[n 2] h[n 2]
0.75[n 5] 0.75h[n 5]
• Hence because of the linearity property we
get
y[n] 0.5h[n 2] 1.5h[n 1]
h[n 2] 0.75h[n 5]
Time-Domain Characterization
of LTI Discrete-Time System
• Now, any arbitrary input sequence x[n] can
be expressed as a linear combination of
delayed and advanced unit sample sequences
in the form
x[n] x[k ] [n k ]
k
• The response of the LTI system to an input
x[k ] [n will
k ] be x[k ] h[n k ]
Time-Domain Characterization
of LTI Discrete-Time System
• Hence, the response y[n] to an input
x[n] x[k ] [n k ]
k
will be
y[n] x[k ] h[n k ]
k
which can be alternately written as
y[n] x[n k ] h[k ]
k
Convolution Sum
• The summation
y[n] x[k ] h[n k ] x[n k ] h[n]
k k
is called the convolution sum of the
sequences x[n] and h[n] and represented
compactly as
y[n] = x[n] * h[n]
Convolution Sum
• Properties -
• Commutative property:
x[n] * h[n] = h[n] * x[n]
• Associative property :
(x[n] * h[n]) * y[n] = x[n] * (h[n] * y[n])
• Distributive property :
x[n] * (h[n] + y[n]) = x[n] * h[n] + x[n] * y[n]
Simple Interconnection
Schemes
• Two simple interconnection schemes are:
• Cascade Connection
• Parallel Connection
Cascade Connection
h1[n] h2[n] h2[n] h1[n]
h3[n]
h4[n]
Simple Interconnection Schemes
h1[n]
h2[n] h1[n]
h3[ n ] h 4[ n ] h 2[ n ] * ( h3[ n ] h 4[ n ])
Simple Interconnection Schemes
h2[n] * h4[n] ( 1 [n] 1 [n 1]) * 2( 12 ) n [n]
2 4
( 12 ) n [n] 12 ( 12 ) n 1[n 1]
( 12 ) n [n] ( 12 ) n [n 1]
( 12 ) n [n] [n]
• Therefore
h[n] [n] 12 [n 1] [n] 12 [n 1] [n] [n]
BIBO Stability Condition of an
LTI Discrete-Time System
• BIBO Stability Condition - A discrete-
time is BIBO stable if the output sequence
{y[n]} remains bounded for all bounded
input sequence {x[n]}
• An LTI discrete-time system is BIBO stable
if and only if its impulse response sequence
{h[n]} is absolutely summable, i.e.
S h[n]
n
BIBO Stability Condition of an
LTI Discrete-Time System
• Proof: Assume h[n] is a real sequence
• Since the input sequence x[n] is bounded
we have
x[n] Bx
• Therefore
y[n] h[k ]x[n k ] h[k ] x[n k ]
k k
B x h[k ] B x S
k
BIBO Stability Condition of an
LTI Discrete-Time System
• Thus, S < implies y[n] B y indicating
that y[n] is also bounded
• To prove the converse, assume y[n] is bounded,
i.e., y[n] B y
• Consider the input given by
y[isn]acausal
1x[nsystem
] 2 x[as
n it1has
] a3causal
x[n 2impulse
] 4 x[n 3]
response
{h[n]} {1 2 3 4 }
Causality Condition of an LTI
Discrete-Time System
• Example - The discrete-time accumulator
defined by
n
y[n] [] [n]
is a causal system as it has a causal impulse
response given by
n
h[n] [] [n]
Causality Condition of an LTI
Discrete-Time System
• Example - The factor-of-2 interpolator
defined by
y[n] xu [n] 1 xu [n 1] xu [n 1]
2
is noncausal as it has a noncausal impulse
response given by
{h[n]} {0.5 1 0.5}
Causality Condition of an LTI
Discrete-Time System
• Note: A noncausal LTI discrete-time system
with a finite-length impulse response can
often be realized as a causal system by
inserting an appropriate amount of delay
• For example, a causal version of the factor-
of-2 interpolator is obtained by delaying the
input by one sample period:
y[n] xu [n 1] 1 xu [n 2] xu [n]
2
Finite-Dimensional LTI
Discrete-Time Systems
• An important subclass of LTI discrete-time
systems is characterized by a linear constant
coefficient difference equation of the form
N M
d k y[n k ] pk x[n k ]
k 0 k 0
• x[n] and y[n] are, respectively, the input and
the output of the system
• {d k } and { pk } are constants characterizing
the system
Finite-Dimensional LTI
Discrete-Time Systems
• The order of the system is given by
max(N,M), which is the order of the difference
equation
• It is possible to implement an LTI system
characterized by a constant coefficient
difference equation as here the computation
involves two finite sums of products
Finite-Dimensional LTI
Discrete-Time Systems
• If we assume the system to be causal, then
the output y[n] can be recursively computed
using
N d M p
y[n] k y[n k ] k x[n k ]
k 1d 0 k 1d 0
provided d 0 0
• y[n] can be computed for all n no ,
knowing x[n] and the initial conditions
y[no 1], y[no 2], ..., y[no N ]
Classification of LTI Discrete-
Time Systems
Based on Impulse Response Length -
• If the impulse response h[n] is of finite
length, i.e.,
h[n] 0 for n N1 and n N 2 , N1 N 2
then it is known as a finite impulse
response (FIR) discrete-time system
• The convolution sum description here is
N2
y[n] h[k ]x[n k ]
k N1
Classification of LTI Discrete-
Time Systems
• The output y[n] of an FIR LTI discrete-time
system can be computed directly from the
convolution sum as it is a finite sum of
products
• Examples of FIR LTI discrete-time systems
are the moving-average system and the
linear interpolators
Classification of LTI Discrete-
Time Systems
• If the impulse response is of infinite length,
then it is known as an infinite impulse
response (IIR) discrete-time system
• The class of IIR systems we are concerned
with in this course are characterized by
linear constant coefficient difference
equations
Classification of LTI Discrete-
Time Systems
• Example - The discrete-time accumulator
defined by
y[n] y[n 1] x[n]
is seen to be an IIR system
Classification of LTI Discrete-
Time Systems
• Example - The familiar numerical integration
formulas that are used to numerically solve
integrals of the form
t
y (t ) x()d
0
can be shown to be characterized by linear
constant coefficient difference equations, and
hence, are examples of IIR systems
Classification of LTI Discrete-
Time Systems
• If we divide the interval of integration into
n equal parts of length T, then the previous
integral can be rewritten as
nT
y (nT ) y ((n 1)T ) x()d
( n 1)T
where we have set t = nT and used the
notation nT
y (nT ) x()d
0
Classification of LTI Discrete-
Time Systems
• Using the trapezoidal method we can write
nT
T {x (( n 1)T ) x ( nT )}
x ( ) d
2
( n 1)T
• Hence, a numerical representation of the
definite integral is given by
m y[m ]x[m] rxy []
• Thus, r [] is obtained by time-reversing
yx
rxy []
Correlation of Signals
• The autocorrelation sequence of x[n] is given
by
rxx [] n x[n]x[n ]
obtained by setting y[n] = x[n] in the
definition of the cross-correlation sequence
rxy []
• Note: r [0] x 2
[ n ] E , the energy
xx n x
of the signal x[n]
Correlation of Signals
• From the relation ryx [] rxy []it follows
that rxx [] rxx [] implying that rxx []is an
even function for real x[n]
• An examination of
rxy [] n x[n] y[n ]
reveals that the expression for the cross-
correlation looks quite similar to that of the
linear convolution
Correlation of Signals
• This similarity is much clearer if we rewrite
the expression for the cross-correlation as
rxy [] n x[n] y[( n)] x[] * y[]
• The cross-correlation of y[n] with the
reference signal x[n] can be computed by
processing x[n] with an LTI discrete-time
system of impulse response y[ n]
x[n ] y[ n ] rxy [n ]
Correlation of Signals
• Likewise, the autocorrelation of x[n] can be
computed by processing x[n] with an LTI
discrete-time system of impulse response
x[ n]
x[n ] x[ n ] rxx [n ]
Correlation Computation
Using MATLAB
• The cross-correlation and autocorrelation
sequences can easily be computed using
MATLAB
• Example - Consider the two finite-length
sequences
x[n] 1 3 2 1 2 1 4 4 2
y[n] 2 1 4 1 2 3
Correlation Computation
Using MATLAB
• The cross-correlation sequence rxy [n]
computed using Program 2_7 of text is
plotted below
30
20
Amplitude
10
-10
-4 -2 0 2 4 6 8
Lag index
Correlation Computation
Using MATLAB
• The autocorrelation sequence rxx []
computed using Program 2_7 is shown below
• Note: At zero lag, rxx [0] is the maximum
60
40
Amplitude
20
-20
-5 0 5
Lag index
Correlation Computation
Using MATLAB
• The plot below shows the cross-correlation
of x[n] and y[n] x[n N ] for N = 4
• Note: The peak of the cross-correlation is
precisely the value of the delay N
60
40
Amplitude
20
-20
-10 -5 0 5
Lag index