Signals and Systems
Lecture 8
Instructor: Sohaib Abdul Rehman
October 21, 2020
Contents
Properties of LTI Systems
Properties of LTI Systems
We will see how the Impulse Response of an LTI system can be
used to determine its following properties:
• Memory of systems
• Invertibility of systems
• Causality of systems
• Stability of systems
Memory of LTI systems - 1
A system is memory less if its output depends on the input at that
time only.
Memory of LTI systems - 1
A system is memory less if its output depends on the input at that
time only.
For an impulse function i.e. δ(t) as input, the output of a
memoryless system is also delta function.
Memory of LTI systems - 1
A system is memory less if its output depends on the input at that
time only.
For an impulse function i.e. δ(t) as input, the output of a
memoryless system is also delta function.
Memory of LTI Systems - 2
Consider a memoryless system with the impulse response h(t).
h(t) = aδ(t)
Then the output of the system will be:
y (t) = x(t) ∗ h(t) = x(t) ∗ δ(t)
Z
= x(τ )aδ(t − tau)dτ = ax(t)
Similarly,
y [n] = x[n] ∗ h[n] = ax[n]
Memory of LTI Systems - 2
Consider a memoryless system with the impulse response h(t).
h(t) = aδ(t)
Then the output of the system will be:
y (t) = x(t) ∗ h(t) = x(t) ∗ δ(t)
Z
= x(τ )aδ(t − tau)dτ = ax(t)
Similarly,
y [n] = x[n] ∗ h[n] = ax[n]
Invertibility of LTI Systems
Consider an LTI system with the impulse response h(t) and its
inverse system h1 (t).
y (t) = h(t) ∗ x(t)
And
w (t) = h1 (t) ∗ y (t) = h1 (t) ∗ h(t) ∗ x(t)
Invertibility of LTI Systems
Consider an LTI system with the impulse response h(t) and its
inverse system h1 (t).
y (t) = h(t) ∗ x(t)
And
w (t) = h1 (t) ∗ y (t) = h1 (t) ∗ h(t) ∗ x(t)
Since h1 (t) is the inverse of h(t), w (t) = x(t)
Invertibility of LTI Systems
Consider an LTI system with the impulse response h(t) and its
inverse system h1 (t).
y (t) = h(t) ∗ x(t)
And
w (t) = h1 (t) ∗ y (t) = h1 (t) ∗ h(t) ∗ x(t)
Since h1 (t) is the inverse of h(t), w (t) = x(t)
=⇒ h(t) ∗ h1 (t) = δ(t)
Invertibility of LTI Systems
Consider an LTI system with the impulse response h(t) and its
inverse system h1 (t).
y (t) = h(t) ∗ x(t)
And
w (t) = h1 (t) ∗ y (t) = h1 (t) ∗ h(t) ∗ x(t)
Since h1 (t) is the inverse of h(t), w (t) = x(t)
=⇒ h(t) ∗ h1 (t) = δ(t)
Similarly
h[n] ∗ h1 [n] = δ[n]
Exercise
Consider the system of the form y (t) = x(t − to )
Determine:
1. the impulse response
2. the inverse function i.e. h1 (t)
3. h(t) ∗ h1 (t)
Causality of LTI Systems
A system is causal if its output at any time depends on:
• the input at that time
• or the input at the past time values
Causality of LTI Systems
A system is causal if its output at any time depends on:
• the input at that time
• or the input at the past time values
For LTI systems, the causality property is satisfied if:
h[n] = 0 for n < 0
Causality of LTI Systems
A system is causal if its output at any time depends on:
• the input at that time
• or the input at the past time values
For LTI systems, the causality property is satisfied if:
h[n] = 0 for n < 0
This is true because the output of LTI systems is the sum of scaled
impulse responses
Causality of LTI Systems
A system is causal if its output at any time depends on:
• the input at that time
• or the input at the past time values
For LTI systems, the causality property is satisfied if:
h[n] = 0 for n < 0
This is true because the output of LTI systems is the sum of scaled
impulse responses
∞
X ∞
X
y [n] = h[k]x[n − k] = h[k]x[n − k]
−∞ 0
Causality of LTI Systems
A system is causal if its output at any time depends on:
• the input at that time
• or the input at the past time values
For LTI systems, the causality property is satisfied if:
h[n] = 0 for n < 0
This is true because the output of LTI systems is the sum of scaled
impulse responses
∞
X ∞
X
y [n] = h[k]x[n − k] = h[k]x[n − k]
−∞ 0
Similarly,
h(t) = 0 for t < 0
And Z ∞
y (t) = h(τ )x(t − τ )dτ
0
Stability of LTI Systems - 1
A system is stable if for every bounded input, it returns a bounded
output.
Stability of LTI Systems - 1
A system is stable if for every bounded input, it returns a bounded
output.
Consider a bounded input x[n] to an LTI system with the impulse
response h[n]
|x[n]| < B
The amplitude of the ouput will be:
∞
X
|y [n]| = |x[n] ∗ h[n]| = h[k]x[n − k]
−∞
Stability of LTI Systems - 1
A system is stable if for every bounded input, it returns a bounded
output.
Consider a bounded input x[n] to an LTI system with the impulse
response h[n]
|x[n]| < B
The amplitude of the ouput will be:
∞
X
|y [n]| = |x[n] ∗ h[n]| = h[k]x[n − k]
−∞
∞
X
|y [n]| = h[k]x[n − k]
−∞
∞
X
≤ |h[k]||x[n − k]|
−∞
This is known as the Triangle Inequality.
Stability of LTI Systems - 2
∞
X
|y [n]| ≤ |h[k]||x[n − k]|
−∞
Since, |x| < B
∞
X
=⇒ |y [n]| ≤ B |h[k]|
−∞
For y [n] to be bounded, we have the following condition:
∞
X
|h[k]| < ∞
−∞
Stability of LTI Systems - 2
∞
X
|y [n]| ≤ |h[k]||x[n − k]|
−∞
Since, |x| < B
∞
X
=⇒ |y [n]| ≤ B |h[k]|
−∞
For y [n] to be bounded, we have the following condition:
∞
X
|h[k]| < ∞
−∞
Similarly, for continuous time systems:
Z ∞
|h(τ )|dτ < ∞
−∞
Exercise
Determine if the h(t) = e −at stable?
Unit Step Response of LTI systems
Consider an LTI with the Impulse response h(t), its response to a
unit step function would be:
Z ∞
s(t) = u(t) ∗ h(t) = h(τ )u(t − τ )dτ
−∞
Unit Step Response of LTI systems
Consider an LTI with the Impulse response h(t), its response to a
unit step function would be:
Z ∞
s(t) = u(t) ∗ h(t) = h(τ )u(t − τ )dτ
−∞
The function u(t − τ ) is 1, for τ ≤ t
Z t
s(t) = h(τ )dτ
−∞
Unit Step Response of LTI systems
Consider an LTI with the Impulse response h(t), its response to a
unit step function would be:
Z ∞
s(t) = u(t) ∗ h(t) = h(τ )u(t − τ )dτ
−∞
The function u(t − τ ) is 1, for τ ≤ t
Z t
s(t) = h(τ )dτ
−∞
Homework: For DT systems, show that the step response is:
n
X
s[n] = h[n]
−∞
