EEE240 Signals and Systems
Chapter 2 – LTI Systems
Ayman Khalil, PhD
Outline
DT LTI Systems : The convolution sum
CT LTI Systems : The convolution integral
Properties of Convolution and LTI Systems
Systems Described by Differential and Difference Equations
and Determining Their Impulse Responses
[2]
A System…
A system is
– Something that transforms signals (hardware or software)
– Representations of physical entities that react to input excitations
– A mathematical function: given x, calculate y
- independent variable = time
- dependent variable = voltage, rate, sound pressure, etc.
Systems can work in continuous time,
or discrete time,
[3]
A System….
[4]
Linear Time Invariant (LTI) Systems
A Linear Time-Invariant System (LTIS) is a system which is
• Linear (properties of homogeneity and additivity)
•Time invariant
Why are LTI systems interesting to us?
• Able to model so many real-world systems
• . . . and can even approximate many non-linear systems with LTI
systems
• We can do transforms on them: Laplace transform, Fourier
transform.
[5]
Linear Time Invariant (LTI) Systems
[6]
Linear Time Invariant (LTI) Systems
[7]
Linear Time Invariant (LTI) Systems
Example: A signal defined as a sum of shifted and scaled impulses
[8]
Convolution
The output of an LTIS can be calculated as the convolution of the input x() with
the impulse response h(). The CT case is just a generalization of the DT case:
[9]
Convolution
[10]
Convolution
Basically, we have an input x(t), we can use convolution to get the output y(t)
from an LTIS defined by h(t):
and in the discrete case,
In other words, the output of a system is the convolution of the input and the
impulse response.
[11]
Convolution
[12]
Convolution
[13]
Convolution
[14]
Convolution
[15]
Convolution
[16]
Convolution
[17]
Convolution
[18]
Convolution
[19]
Properties of Convolution
[20]
Properties of Convolution
Memory property in LTI systems:
An LTI system is memoryless if and only if
Causality property in LTI systems
An LTI system is causal if and only if
Stability property in LTI systems
An LTI system is stable if and only if its impulse response is absolutely summable
(integrable)
Invertibility property in LTI systems
If an LTI system is invertible then it has an LTI inverse system, which if connected in
cascade with the first system produces an overall identity system.
[21]
Systems Described by Differential and Difference Equations and
Determining Their Impulse Responses
Many systems are described by differential or difference equations.
We will consider only Linear Constant Coefficient Differential (Difference) equations
(LCCDE)..
Consider a simple differential equation : The solution contains a particular and a
homogeneous part :
• The differential equation by itself does not specify a unique solution.
• An auxiliary (boundary) condition is required to find a unique solution.
• Different auxiliary conditions can lead to different solutions.
• Not all auxiliary conditions lead to LTI systems.
•In this course, to obtain causal and LTI systems from differential/difference equations,
the initial rest conditions will be used.
[22]
(Unit) step response to an LTIS
In practice, impulses are difficult to generate and work with, so it can be difficult to
study an LTIS with impulse responses. We're often interested in the response of an LTIS
to the unit step
Like if we step up the temperature in an air-conditioning system, the voltage in an
electrical circuit, the dosage of a drug to a patient, etc.
[23]
Linear Constant-Coefficient Differential Equations
First-order differential equation
where y(t) denotes the output of the system and x(t) is the input.
Example: consider the input
where K is a real number.
The complete solution consists of the sum of a particular solution yp(t),
and a homogeneous solution, yh(t),
where the particular solution satisfies the main equation and yh (t) is a solution of the
homogeneous differential equation
[24]
Linear Constant-Coefficient Differential Equations
A common method for finding the particular solution for an exponential input signal as
in the main equation is to look for a so-called forced response-i.e., a signal of the same
form as the input.
where Y is a number that we must determine.
[25]
Linear Constant-Coefficient Differential Equations
In order to determine yh (t), we hypothesize a solution of the form
From this equation, we see that we must take s= -2
In order for the value of A to be determined, we need to specify an auxiliary condition the
condition of initial rest implies that y(t) = 0 for t < 0.
y(0) = 0 yields
[26]
linear Constant-Coefficient Difference Equations
Consider the difference equation
[27]
linear Constant-Coefficient Difference Equations
Highlighting the fact that we need the previous value of the output, y[n- 1], to calculate the
current value. Thus, to begin the recursion, we need an initial condition. For example,
suppose that we impose the condition of initial rest and consider the input
In this case, since x[ n] = 0 for n<=- 1, the condition of initial rest implies that y[ n] = 0 for n
<= - 1, so that we have as an initial condition y[ -1] = 0. Starting from this
initial condition, we can solve for successive values of y[n] for n >= 0 as follows:
[28]
Block Diagram Representations of First-Order Systems
Described by Differential and Difference Equations
[29]
