Signal and Systems Lecture 2 Department of ECE
Signal and Systems Lecture 2
Department of ECE
Solomon T. Mawcha
Address:
solomit123@gmail.com
Mekelle University-Mekelle Institute of Technology
April 4, 2017
1/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Outline
1 Linear Time-Invariant Systems
2 Convolution of Discrete-time LTI-system
3 Continuous-time LTI System
4 Properties of Convolution
5 LTI System Properties
1/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Introduction
a LTI system is a system which is
1 linear
2 time-invariant
why LTI systems are interesting to us?
1 able to model so many real world systems
2 can even approximate many non-linear systems
3 we can do transforms on them: LT and FT
h[n]/h(t) for discrete-time/Continuous-time signal
1 is impulse response of an LTI system S
2 is the response of the system to a unit impulse input
2/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Important property of the Unit impulse
let’s recall the definition of an unit impulse
1
(
1 if n = 0
δ[n] =
0 if otherwise
x[n] can be represented as linear combination of delayed
unit impulses
representation of a signal in terms of impulses
1 X [n]δ[n] = X [0]δ[n] = X [0]forn = 0 otherwise 0
2 X [n]δ[n + 1] = X [−1]δ[n + 1] = X [−1]forn = −1 otherwise 0
3 X [n]δ[n − 1] = X [1]δ[n − 1] = X [1]forn = 1 otherwise 0
2/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Representation of Discrete signals using impulses
sum of similar sequences can be represented
1 X [n] = ...X [−3]δ[n + 3] + X [−2]δ[n + 2] + X [−1]δ[n + 1] +
X [0]δ[n]P+ X [1]δ[n − 1] + X [2]δ[n − 2] + X [3]δ[n − 3] + ...
∞
2 X [n] = k=−∞ x[k]δ[n − k ]
3 this is a linear combination of shifted unit impulses of
δ[n − k]
2/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Figure 1 : A general discrete- time signal expressed as a
superposition of weighted, delayed unit impulses
2/7
Signal and Systems Lecture 2 Department of ECE
Linear Time-Invariant Systems
Representation of Continuous signals using
impulses
1 x(t)δ(t) = x(0)δ(t) for t = 0
R∞
−∞ δ(t)x(t)dt = x(0)
2
R∞
3 Time shifting −∞ δ(t − t0 )x(t)dt = x(t0 )
2/7
Signal and Systems Lecture 2 Department of ECE
Convolution of Discrete-time LTI-system
Outline
1 Linear Time-Invariant Systems
2 Convolution of Discrete-time LTI-system
3 Continuous-time LTI System
4 Properties of Convolution
5 LTI System Properties
2/7
Signal and Systems Lecture 2 Department of ECE
Convolution of Discrete-time LTI-system
LTI system Convolution sum
convolution determines the system’s output from
knowledge of the input and the system’s impulse response
y[n] = ∞
P
k=−∞ x[k ]h[n − k ]
if we are analyzing a given system (e.g., a circuit) we may
need to compute a convolution to determine how it
behaves in response to various different input signals
y[n] = x[n] ∗ h[n] ∗ is convolution operator not
multiplication operator
if m and P are lengths of x[k ]andh[n − k], then length of
y[n] is m + p − 1
3/7
Signal and Systems Lecture 2 Department of ECE
Convolution of Discrete-time LTI-system
Example 1
Consider an input x[n] and a unit impulse h[n] are given by
x[n] = αn u[n] where 0 < α < 1
h[n] = u[n]
Figure 1 : The signals x[n] and h[n]
3/7
Signal and Systems Lecture 2 Department of ECE
Convolution of Discrete-time LTI-system
Solution
y [n] = x[n] ∗ h[n] = ∞
P
k=−∞ x[k]x[n − k]
x[k]x[n − k] = 0 for P
n<0
y[n] = x[n] ∗ h[n] = nk=0 αk for n ≥ 0
1
y[n] = 1−α (1 − αn+1 )u[n]
Figure 1 : Output y [n]
3/7
Signal and Systems Lecture 2 Department of ECE
Convolution of Discrete-time LTI-system
Example 2
Figure 1 : Output y [n]
3/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Outline
1 Linear Time-Invariant Systems
2 Convolution of Discrete-time LTI-system
3 Continuous-time LTI System
4 Properties of Convolution
5 LTI System Properties
3/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
The Convolution Integral
convolution determines the system’s output from
knowledge of the input and the system’s impulse response
R ∞ by y(t) = x(t) ∗ h(t)
represented
y(t) = −∞ x(τ )h(t − τ )dτ
Width
1 In continuous time, if Duration x1 = T1 and Duration
x2 = T1 , then
2 Duration x1 (t) ∗ x2 (t) = T 1 + T 2
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Intuitive explanation of Convolution
assume the impulse response decays linearly from t = 0 to
zero at t = 1
divide input x(τ ) into pulses.
the system response at t is then determined by x(τ )
weighted by h(t − τ )
(i.e. x(τ )h(t − τ )) for the shaded pulse, PLUS the
contribution from all the previous pulses of x(τ )
The summation of all these weighted inputs is the
convolution integral
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Figure 2 : Intuitive explanation of Convolution.
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Steps for Graphical Convolution of x(t) ∗ h(t)
R∞
y (t) = −∞ x(τ )h(t − τ )dτ
rewrite the signals as functions of τ , x(τ ) and h(τ )
flip just one of the signals around t = 0 to get either x(−τ )
or h(−τ )
1 it is usually best to flip the signal with shorter duration
2 for notational purposes here: we’ll flip h(τ ) to get h(−τ )
shift h(−τ ) by an arbitrary value of t to get h(t − τ ) and get
its edges
find Regions of τ -Overlap
1 what we are trying to do here is find intervals of t over which
the product x(τ )h(t − τ ) has a single mathematical form in
terms of τ
2 in each region find: Interval of t that makes the identified
overlap happen
3 working examples is the best way to learn how this is done
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Steps continued
for each Region: form the Product x(τ )h(t − τ ) and
Integrate
1 form product, find the Limits of integration and integrate the
product
assemble the output from the output time-sections for all
the regions
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Example 1: CT signal Convolution
Determine graphically y (t) = x(t) ∗ h(t) for x(t) = e−t u(t) and
h(t) = e−2t u(t)
Figure 2 : Calculation of the convolution integral using graphical
method
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Figure 2 : Calculation of the convolution integral using graphical
method
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Figure 2 : Calculation of the convolution integral using graphical
method
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Figure 2 : Calculation of the convolution integral using graphical
method
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Figure 2 : Calculation of the convolution integral using graphical
method
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Rt
y(t) = o x(τ )h(t − τ )dτ
Rt −τ e−2(t−τ ) dτ
y(t) = 0e
Rt
y(t) = e−2t 0 eτ dτ
y(t) = (e−t − e−2t )u(t)
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Example 2: CT signal Convolution
Let x(t) be the input to an LIT system with unit impulse
response h(t) where
x(t) = e−at u(t) and h(t) = u(t) for a > 0
Figure 2 : Calculation of the convolution integral
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Solution
R∞
y (t) = −∞ x(τ )h(t − τ )dτ
R∞
y (t) = −∞ e−aτ u(τ )u(t − τ )dτ
y (t) = − a1 (e−at − 1)
y (t) = a1 (1 − e−at )u(t)
Figure 2 : Response of the System
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Example 2: CT signal Convolution
Consider the convolution of the following signals:
(
1 if 0 ≤ t ≤ T
x(t) =
0 otherwise
(
t if 0 ≤ t ≤ 2T
h(t) =
0 otherwise
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Solution
Figure 2 : Signal x(τ ) and h(t − τ ) for different value of t 4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Solution
0 if t < 0
12 t 2 if 0 ≤ t ≤ T
y(t) = tT − 21 T 2 if T ≤ t ≤ 2T
tT − 21 t 2 + 32 T 2 if 2T ≤ t ≤ 3T
0 otherwise
4/7
Signal and Systems Lecture 2 Department of ECE
Continuous-time LTI System
Solution
Figure 2 : Signal y(t) = x(t) ∗ h(t) for different value of t 4/7
Signal and Systems Lecture 2 Department of ECE
Properties of Convolution
Outline
1 Linear Time-Invariant Systems
2 Convolution of Discrete-time LTI-system
3 Continuous-time LTI System
4 Properties of Convolution
5 LTI System Properties
4/7
Signal and Systems Lecture 2 Department of ECE
Properties of Convolution
The commutative property
h(t) ∗ x(t) = x(t) ∗ h(t)
1 t 0 = t − τ, t =Rt 0 + τ, τ = t − t 0 , dτ =R−dt 0 ,
∞ ∞
2 x(t) ∗ h(t) = −∞ x(τ )h(t − τ )dτ = −∞ x(t − t 0 )h(t 0 )dt 0 =
h(t) ∗ x(t)
h[n] ∗ x[n] = x[n] ∗ h[n])
let r = n − k Pthen k = n − r
∞
x[n]
P∞ ∗ h[n] = k=−∞ x[k ] ∗ h[n − k ] =
r =−∞ x[n − r ] ∗ h[r ] = h[n] ∗ x[n]
5/7
Signal and Systems Lecture 2 Department of ECE
Properties of Convolution
The associative property
1 (x(t) ∗ h1 (t)) ∗ h2 (t) = x(t) ∗ (h1 (t) ∗ h2 (t))
2 (x[n] ∗ h1 [n]) ∗ h2 [n]= x[n] ∗ (h1 [n] ∗ h2 [n])
Figure 3 : Graphical implication of the associative property of
convolution.
5/7
Signal and Systems Lecture 2 Department of ECE
Properties of Convolution
The distributive property
x(t) ∗ (h1 (t) + h2 (t)) = x(t) ∗ h1 (t) + x(t) ∗ h2 (t)
Figure 3 : The distributive property
5/7
Signal and Systems Lecture 2 Department of ECE
Properties of Convolution
The special property
represented by δ(t) = δ(t) ∗ δ(t)
R∞
1 δ(t) = −∞
δ(τ )δ(t − τ )dτ
x(t) = x(t) ∗ δ(t)
R∞
1 δ(t − t 0 )dt = 1
R−∞
∞ R∞
2
−∞
x(t 0 )δ(t − t 0 )dt 0 == x(t) −∞ δ(t − t 0 )dt 0 = x(t)
δ(t − t1 ) ∗ δ(t − t2 ) = δ(t − t1 − t2 )
R∞
1 δ(t − t1 ) ∗ δ(t − t2 ) = −∞ δ(τ − t1 )δ(t − τ − t2 )dτ
R∞
2 δ(t − t1 ) ∗ δ(t − t2 ) = δ(t − t1 − t2 ) −∞ δ(τ − t1 )dτ
3 δ(t − t1 ) ∗ δ(t − t2 ) = δ(t − t1 − t2 ).1
4 δ(t − t1 ) ∗ δ(t − t2 ) = δ(t − t1 − t2 )
5/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Outline
1 Linear Time-Invariant Systems
2 Convolution of Discrete-time LTI-system
3 Continuous-time LTI System
4 Properties of Convolution
5 LTI System Properties
5/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
LTI System with memory/out Memory
Recap
A system is memoryless if its output at any time depends only
on the value of the input at that the same time.
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Discrete-time LTI systems with/out memory
is memoryless if h[n] = 0 for n not equal zero
the impulse response has the form
1 h[n] = K δ[n]
2 where k = h[0] which is constant
the convolution with X [n] will be
1 y [n] = x[n] ∗ h[n] = x[n] ∗ K δ[n] = k(x[n] ∗ δ[n]) = Kx[n]
2 where k = h[0] which is constant
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Continous-time LTI systems with/out memory
is memoryless if h(t) = 0 for t not equal zero
the impulse response has the form
1 h(t) = K δ(t)
2 where k δ(t) = h(0) which is constant
the convolution with X (t) will be
1 y (t) = x(t) ∗ h(t) = x(t) ∗ K δ(t) = k (x(t) ∗ δ(t)) = Kx(t)
2 if k = 1 y (t) = x(t) the system becomes Identity
3 the input equals with the output
4 the impulse response equal with δ(t)
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Invertibility of LTI systems
Recap:
a system is invertible only if an inverse system exists
connect the inverse system in series with the original
system
1 this produces an output equal to the input of the first system
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Invertible LTI systems
if LTI system is invertible, then it has an LTI inverse
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Causality of LTI systems
Recap:
the output of a causal system depends only on the present
and past values of the input to the system
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Causality of Discrete-time LTI systems
Y [n] must not depend on x[n] for k > n
1 all the coefficients h[n − k ] that multiply values of x[k ] for
n < k must be zero
then the impulse response of a causal discrete LTI system
satisfies the condition
1 h[n] = 0 for n < 0
y[n] = ∞
P P∞
k=−∞ x[k ]h[n − k ] = k=−∞ h[k ]x[n − k ] ...
Commutative property
the convolution sum for LTI discrete Causal system is given
by
P∞
1 y [n] = k=0 h[k]x[n − k ]
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Causality of Continuous-time LTI systems
then the impulse response of a causal continuous LTI
system satisfies the condition
1 h(t) = 0 for t < 0
R∞ R∞
y(t) = −∞ x(τ )h(t − τ )dτ = −∞ h(τ )x(t − τ )dτ ...
Commutative property
the convolution integral for continuous LTI Causal system
is given by
R∞
1 y (t) = t=0
h(τ )x(t − τ )dτ
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Stability of LTI systems
a system is stable if every bounded input produces a
bounded output
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Stability of Discrete LTI systems
assume x[n] < B for for all n
supply the input to an LTI system with impulse response
h[n]
P∞
1 y [n] = k=−∞ h[k ]x[n − k ]
P∞
2 |y [n]| = k=−∞ |h[k ]x[n − k]|
P∞
3 |y [n]| ≤ k=−∞ |h[k ]||x[n − k]|
P∞
4 |y [n]| ≤ B k =−∞ |h[k ]| for all value of k and n
if the impulse response is absolutely summable then y[n]
is bounded in magnitude... stable
P∞
1 that is if k=−∞ |h[k]| < ∞
2 |y [n]| < ∞ which is stable
6/7
Signal and Systems Lecture 2 Department of ECE
LTI System Properties
Stability of Continuous LTI systems
assume x(t) < B for for all t
supply the input to an LTI system with impulse response
h(t)
R∞
1 y (t) = −∞ h(τ )x(t − τ )dτ
R∞
2 |y (t)| = −∞ |h(τ )x(t − τ )|dτ
R∞
3 |y (t)| ≤ −∞ |h(τ )||x(t − τ )|dτ
R∞
4 |y (t)| ≤ B −∞ |h(τ )|dτ for all value of k and n
if the impulse response is absolutely integrable, then y(t)
is bounded in magnitude... stable
R∞
1 that is if −∞ |h(τ )|dτ < ∞
2 |y (t)| < ∞ which is stable
6/7
Signal and Systems Lecture 2 Department of ECE
”The enemies of people are those who keep them in
ignorance”
Thomas Sankara
7/7
Signal and Systems Lecture 2 Department of ECE
Vielen Dank!
7/7
