EECE 301

Signals & Systems

Prof. Mark Fowler
Note Set #8
D-T Convolution: The Tool for Finding the
Zero-State Response
Reading Assignment: Section 2.1-2.2 of Kamen and Heck

1/14

Course Flow Diagram

The arrows here show conceptual flow between ideas. Note the parallel structure between
the pink blocks (C-T Freq. Analysis) and the blue blocks (D-T Freq. Analysis).
New Signal
Models
Ch. 1 Intro
C-T Signal Model
Functions on Real Line
System Properties
LTI
Causal
Etc
D-T Signal Model
Functions on Integers

New Signal
Model
Powerful
Analysis Tool

Ch. 3: CT Fourier
Signal Models

Ch. 5: CT Fourier
System Models

Ch. 6 & 8: Laplace

Models for CT
Signals & Systems

Fourier Series
Periodic Signals
Fourier Transform (CTFT)
Non-Periodic Signals

Frequency Response
Based on Fourier Transform

Transfer Function

New System Model

New System Model

Ch. 2 Diff Eqs

C-T System Model
Differential Equations
D-T Signal Model
Difference Equations

Ch. 2 Convolution

Zero-State Response

C-T System Model

Convolution Integral

Zero-Input Response
Characteristic Eq.

D-T System Model

Convolution Sum

Ch. 4: DT Fourier
Signal Models
DTFT
(for Hand Analysis)
DFT & FFT
(for Computer Analysis)

Ch. 5: DT Fourier
System Models
Freq. Response for DT
Based on DTFT

New System Model

New System Model

Ch. 7: Z Trans.
Models for DT
Signals & Systems
Transfer Function

New System
Model2/14

Convolution
Our Interest: Finding the output of LTI systems (D-T & C-T cases)
Notice the parallel
structure between C-T
and D-T systems!
Well see that they are
solved using similar
but slightly different
tools!!!

LTI System
C-T
Differential
Equation
(solve)
zero
input
solution
Use char.
poly. roots

D-T
Difference
Equation
(solve)

zero
state
solution

zero
input
solution

HOW??

Use char.
poly. roots

zero
state
solution
HOW??

Our focus in this chapter will be on finding the zero-state

solution (we already know how to find the zero-input solution for C-T
differential equations and later well learn how to do that for D-T
difference equations)

3/14

How do we find the Zero-State Response?

(Remember that is the response (i.e., output) of the system to a specific
input when the system has zero initial conditions)

Recall that in the examples for differential equations we always

saw:
t

y ZS (t ) = h(t ) x( )d

Where does this come from?

t0

How do we deal with it?

C-T convolution

Recall that in the examples for difference equations we saw:

n

y ZS [n] = h[n i ]x[i ]

i =1

Where does this come from?

How do we deal with it?

D-T convolution

Well handle D-T systems first because they are easier to understand!
4/14

Convolution for LTI D-T systems

We are trying to find yZS(t) so the ICs = 0

i.e. no stored energy

Well drop the zs subscript to make the notation easier!

x[n]

LTI
D-T system
ICs = 0

y[n]

Before we can find the Z-S outptut we need something first:

Impulse Response (Warning: book calls it unit-pulse response)

The impulse response h[n] is what comes out when [n] goes in w/ ICs=0
[n]

h[n]

[n]

LTI
D-T system
ICs = 0

n
h[n]
Note: If system is causal,
then h[n] = 0 for n < 0 5/14

The impulse response h[n] uniquely describes the system so we

can identify the system by specifying its impulse response h[n].
Thus, we often show the system using a block diagram with the
systems impulse response h[n] inside the box representing the
system:
x[n]

LTI
D-T system
with h[n]

y[n]

Because impulse response is only defined for LTI systems, if

you see a box with the symbol h[n] inside it you can assume that
the system is an LTI system.
x[n]

h[n]

y[n]

6/14

How do we know/get the impulse response h[n]?

Many possible ways:
1. Given by the designer of D-T systems
2. Measured experimentally
-Put in sequence . . . 0 0 1 0 0 0 . . .
-See what comes out
3. Mathematically analyze the D-T system
-Given difference equation
-Derive h[n]

There are many

ways to do this,
as we will see!

In what form will we know h[n]?

1. h[n] known analytically as a function
Ex:

1
h[n] = u[n]
2

2. h[n] known numerically as a finite-duration sequence

Ex:

n 0
1
2
3 4 5
h[n] 0.5 1 2.1 1.3 .6 0 . . .

We assume that
h[n] = 0 for n < 0
7/14

Example of analytically finding h[n]

Given a system described by a 1st order difference equation:

y[n] = ay[n 1] + bx[n]

(a and b are arbitrary numbers)

Recall that h[n] is what comes out when [n] goes in (with zero ICs).
So we can re-write the above difference equation as follows:

h[n ] = ah[n 1] + b [n ]
Here we solve for h[n] recursively and then examine the results to deduce a
closed-form solution (note: we cant always use this deductive approach):
n
[n]
h[n]
-1
0
a 0+ b0 = 0
By examining these
a 0 + b 1 = b
0
1
results we see
ab + b 0 = ab
1
0
n
2
=
b
(

a
)
u[n]
a (ab) = (a) b
2
0
3
0
a ( a ) 2 b = (a) 3 b

h[n ] = b( a ) n u[n ]

So we now have the impulse

response for this system!!! Next
well learn how to use it to solve for
the zero-state response!!!
8/14

Q: How do we use h[n] to find the Zero-State Response?

A: Convolution Well go through three analysis steps that will derive
The General Answer that convolution is what we need to do to find the zerostate response
After that we wont need to re-do these steps well just Do Convolution
Step 1: Using time-invariance we know:

[n-i]

h[n]

h[n-i]

(w/ ICs = 0)
Shifted input gives
shifted output

Step 2: Use homogeneity part of linearity: x[i][n-i]

h[n]

x[i]h[n-i]

(w/ ICs = 0)
The input is a function of n
so we view x[i] as a fixed
number for a given i

So we scale the output

by the same fixed number
9/14

Lets see step 2

for a specific input:
x[i][n-i]

h[n]

x[i]h[n-i]

x[i]
3
2
1

-1

(w/ ICs = 0)

i
1 2 3 4 5 6
x[0][n]

This In
n

This Out
1h[n]

h[n]
ICs = 0

x[1][n-1]
n

2h[n-1]
h[n]
ICs = 0

x[2][n-2]

2.5

1h[n-2]
h[n]
ICs = 0

x[3][n-3]
n

2.5h[n-3]
h[n]
ICs = 0
10/14

Step 3: Use additivity part of linearity

In Step 2 we looked at inputs like this:
x[i][n-i]

h[n]
ICs = 0

For each i, a different input

x[i]h[n-i]

For each i, a different response

Now we use the additivity part of linearity:

Put the Sum of Those Inputs In

Get the Sum of Their Responses Out

Input:

x[i ] [n i ]

i =

Output:

x[i ]h[n i ]

i =

But what is this??

On the next slide we show that it is
the desired input signal x[n]!
11/14

Lets see step 3 for

a specific input:

x[i ] [n i ]

3
2
1
-1

x[i]
i
1 2 3 4 5 6

i =

x[0][n]

1h[n]
h[n]
ICs = 0

Note: The
Sum of these
x-weighted
impulses
gives x[n]!!

x[1][n-1]
n

2h[n-1]
h[n]
ICs = 0

x[2][n-2]

2.5

1h[n-2]
h[n]
ICs = 0

x[3][n-3]
n

2.5h[n-3]
h[n]
ICs = 0
12/14

So what weve seen is this:

Input:

x[i ] [n i ]

Output:

x[i ]h[n i ]

i =

i =

= x[n]
Or in other words weve derived an expression that tells
what comes out of a D-T LTI system with input x[n]:

x[n]

y[ n ] =
h[n]
ICs = 0

x[i ]h[n i ]

CONVOLUTION!

i =

y[n] = x[n] h[n]

Notation for Convolution

So now that we have derived this result we dont have to do these

three steps we just use this equation to find the zero-state output:

y[ n ] =

x[i ]h[n i ]

i =

CONVOLUTION!
13/14

Big Picture
For a LTI D-T system in zero state we no longer need the
difference equation model
-Instead we need the impulse response h[n] & convolution
New alternative model!

Difference
Equation
Equivalent Models (for zero state)
Convolution
& Impulse resp.
14/14