6.

003: Signals and Systems

Z Transform

February 23, 2010

Mid-term Examination #1

Wednesday, March 3, 7:30-9:30pm. No recitations on the day of the exam. Coverage: Representations of CT and DT Systems Lectures 17 Recitations 18 Homeworks 14

Homework 4 will not collected or graded. Solutions will be posted.

Closed book: 1 page of notes (8
1
11 inches; front and back).
2
Designed as 1-hour exam; two hours to complete. Review sessions during open oce hours.

Z Transform

Z transform is discrete-time analog of Laplace transform.

Z Transform

Z transform is discrete-time analog of Laplace transform.

Furthermore, you already know about Z transforms (we just havent called them Z transforms) ! Example: Fibonacci system dierence equation operator expression system functional unit-sample response y[n] = x[n] + y[n1] + y[n2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

Check Yourself

Example: Fibonacci system dierence equation operator expression system functional unit-sample response y[n] = x[n] + y[n1] + y[n2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

What is the relation between system functional and h[n]?

Check Yourself

system functional unit-sample response Expand functional in a series:

1 +R +2R2 +3R3 +5R4 +8R5 + 1 R R2 1 1 R R2 R +R2 R R2 R3 2R2 +R3 2R2 2R3 2R4 3R3 +2R4 3R3 3R4 3R5

Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

Y 1 = 1 + R + 2R2 + 3R3 + 5R4 + 8R5 + 13R6 + = X 1 R R2 = h[0] + h[1]R + h[2]R2 + h[3]R3 + h[4]R4 + h[n]Rn =
n

Check Yourself

Example: Fibonacci system dierence equation operator expression system functional unit-sample response y[n] = x[n] + y[n 1] + y[n 2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

What is the relation between system functional and h[n]? Y = h[n]Rn X n

Check Yourself

Example: Fibonacci system dierence equation operator expression system functional unit-sample response Y = h[n]Rn X n Whats the relation between H(z) and h[n]? y[n] = x[n] + y[n 1] + y[n 2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

Check Yourself

Series expansion of system functional:

Y h[n]Rn = X n 1 Substitute R : z H(z) =
n

h[n]z n

Check Yourself

Example: Fibonacci system dierence equation operator expression system functional unit-sample response Y = h[n]Rn X n Whats the relation between H(z) and h[n]? H(z) =
n

y[n] = x[n] + y[n 1] + y[n 2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

h[n]z n

Concept Map: Discrete-Time Systems

Multiple representations of DT systems.

Delay R

Block Diagram
X +
Delay

System Functional
Y

+
Delay

Y 1 = X 1 R R2

Unit-Sample Response

index shift

h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, . . .

Dierence Equation

y[n] = x[n] + y[n1] + y[n2]

System Function
Y (z) z2 H(z) = = X(z) 1 z z2

Concept Map: Discrete-Time Systems

Relation between Unit-Sample Response and System Functional.

Delay R

Block Diagram
X +
Delay

System Functional
Y

+
Delay

Y 1 = X 1 R R2
series

Unit-Sample Response

index shift

h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, . . .

Dierence Equation

y[n] = x[n] + y[n1] + y[n2]

System Function
Y (z) z2 H(z) = = X(z) 1 z z2

Concept Map: Discrete-Time Systems

Relation between System Functional and System Function.

Delay R

Block Diagram
X +
Delay

System Functional
Y

+
Delay

Y 1 = X 1 R R2

Unit-Sample Response
index shift

h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, . . .

1 R z

Dierence Equation

y[n] = x[n] + y[n1] + y[n2]

System Function
Y (z) z2 H(z) = = X(z) 1 z z2

Concept Map: Discrete-Time Systems

Relation between Unit-Sample Response and System Function.

Delay R

Block Diagram
X +
Delay

System Functional
Y

+
Delay

Y 1 = X 1 R R2
series

Unit-Sample Response
index shift

h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, . . .

1 R z

Z transform Dierence Equation System Function

Y (z) z2 H(z) = = X(z) 1 z z2

y[n] = x[n] + y[n1] + y[n2]

Check Yourself

Example: Fibonacci system dierence equation operator expression system functional unit-sample response Y = h[n]Rn X n H(z) =
n

y[n] = x[n] + y[n 1] + y[n 2] Y = X + RY + R2 Y Y 1 = X 1 R R2 h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, . . .

h[n]z n

Z transform!

Z Transform

Z transform is discrete-time analog of Laplace transform.

Z transform maps a function of discrete time n to a function of z.
X(z) =
n

x[n]z n

There are two important variants: Unilateral X(z) = Bilateral X(z) =

n= n=0

x[n]z n

x[n]z n

Dierences are analogous to those for the Laplace transform.

Check Yourself

Find the Z transform of the unit-sample signal.

[n]

Check Yourself

Find the Z transform of the unit-sample signal.

[n]

n
x[n] = [n] X(z) =
n=

x[n]z n = x[0]z 0 = 1

Z{[n]} = 1, analogous to L{(t)} = 1.

Check Yourself

Find the Z transform of a delayed unit-sample signal.

x[n]

Check Yourself

Find the Z transform of a delayed unit-sample signal.

x[n]

n
x[n] = [n 1] X(z) =
n=

x[n]z n = x[1]z 1 = z 1

Z Transforms

Example: Find the Z transform of the following signal.

n 7 x[n] = 8 u[n]

4321 0 1 2 3 4

n 7 7 n 1 z n n = X(z) = z u[n] = z = 7 z 1 8 8 1 8 z 7
8
n= n=0 7 1 provided 8 z < 1, i.e., |z| > 7 .
8

Z Transforms

Example: Find the Z transform of the following signal.

n
7 x[n] =
8
u[n]

4321 0 1 2 3 4

7 n 7 n 1 z
n n X(z) =
z
u[n] =
z
= = 7 z
1
8

8

1

8
z
7

8
n=
n=0


7 1
provided

8
z

< 1
i.e., |z| >
7
.
, 8

n
7 x[n] =
8
u[n]

ROC

z -plane

z
7
8
4321 0 1 2 3 4
n

7 8

Shape of ROC

Regions of converge for Z transform are delimited by circles. Example: x[n] = n u[n] X(z) =
n=

n u[n]z n =

n z n

1 = ; 1 z 1 z = ; |z| > || z

z 1 < 1

n=0

ROC

z -plane

x[n] = n u[n]

z
z
n

4321 0 1 2 3 4

Shape of ROC

Regions of converge for Laplace transform delimited by vertical lines. Example: x(t) = et u(t) X(s) =

et u(t)est dt =

et est dt

1 = ; s

Re(s) > Re()

x(t) = et u(t)
1 s

s-plane
ROC

Distinguishing Features of Transforms

Most-important feature of Laplace transforms is the derivative rule:

x(t) X(s)
x(t) sX(s)
allows us to use Laplace transforms to solve dierential equations.
Similarly, most-important feature of Z transforms is the delay rule:
x[n] X(z) x[n 1] z 1 X(z) allows us to use Z transforms to solve dierence equations.

Distinguishing Features of Transforms

Delay property x[n] X(z) x[n 1] z 1 X(z) Proof: X(z) =

x[n]z n

n=

Let y[n] = x[n 1] then

Y (z) = y[n]z n = x[n 1]z n
n= n=

Substitute m = n 1 x[m]z m1 = z 1 X(z) Y (z) =

m=

Check Yourself

What DT signal has the following Z transform?

z -plane 7 ; |z| < 7 8 z8 z

ROC
7 8

Check Yourself

If 7 Y (z) =
; |z| <
7 8
z
8

z then y[n] corresponds to the unit-sample response of
Y (z) z H(z) = .
= 7 X(z) z

The dierence equation for this system is

7 y[n + 1] y[n] = x[n + 1] . 8 Convergence inside |z| =
7
corresponds to a left-sided (non-causal)

8
response. Solve by iterating backwards in time:
8 y[n] = (y[n + 1] x[n + 1])
7

Check Yourself

Solve by iterating backwards in time:

8
y[n] = (y[n + 1] x[n + 1])
7
Start at rest:
n x[n] y[n]
>0 0 0
0 1 0

8
1
0
7 8 2 2 0
7 8 3
3 0
7
8 n
n
7

n n 8 7 y[n] = ; n<0= u[1 n]

7 8

Check Yourself

Plot

n
7 y[n] =
8
u[1 n]
4 3 2 1 0 1 2 n z

z -plane
ROC
7 8

z
7
8

Check Yourself

What DT signal has the following Z transform? z -plane

7 ; |z| < 7 8 z8 z

ROC
7 8

y[n] = 8 4 3 2 1

n 7

u[1 n] 1 2 n z

z -plane
ROC
7 8

z7 8

Check Yourself

Find the inverse transform of X(z) = 3z 2z 2 5z + 2

given that the ROC includes the unit circle.

Check Yourself

Find the inverse transform of 3z X(z) = 2 2z 5z + 2 given that the ROC includes the unit circle.
Expand with partial fractions: 3z 1 2 X(z) = 2 = 2z 1 z 2 2z 5z + 2 Not at standard form!

Check Yourself

Standard forms:

n
7 x[n] =
8
u[n]

ROC

z -plane

z
7
8
4321 0 1 2 3 4
n

7 8

n
7
y[n] =
8
u[1 n]


4 3 2 1 0 1 2 n z

z -plane
ROC
7 8

z 7
8

Check Yourself

Find the inverse transform of 3z X(z) = 2 2z 5z + 2 given that the ROC includes the unit circle. Expand with partial fractions: 3z 2 1 X(z) = 2 = 2z 1 z 2 2z 5z + 2 Not at standard form! Expand it dierently: as a standard form: 3z 2z z z z = X(z) = 2 = 1 z2 2z 1 z 2 2z 5z + 2 z2 Standard form: a pole at 1 and a pole at 2. 2

Check Yourself

Ratio of polynomials in z: z z 3z =
X(z) =
2 1
z 2
2

5z + 2 z z
2


a pole at
1
and a pole at 2

. 2

z -plane

ROC

1 2

. Region of convergence is outside pole at

1
but inside pole at 2

2
n 1 x[n] = u[n] + 2n u[1 n]
2

Check Yourself

Plot. n 1 x[n] = u[n] + 2n u[1 n]

2

x[n]

Check Yourself

Alternatively, stick with non-standard form:

3z 1 2 X(z) = 2 = 2z 1 z 2 2z 5z + 2 Make it look more standard: 1 1 z 1 z X(z) = z 1 2z 2 z2 z2

Check Yourself

Alternatively, stick with non-standard form:

3z 1 2
= X(z) = 2 2z 1 z 2
2z 5z + 2 Make it look more standard: 1 1 z 1 z X(z) = z 1 2z z 2
2 z2 Now n 1 1 u[n] + 2R {+2n u[1 n]} x[n] = R 2 2 n1 1 1 = u[n 1] + 2 +2n1 u[n] 2 2 n 1 = u[n 1] + {+2n u[n]}
2 x[n]

Check Yourself

Alternative 3: expand as polynomials in z 1 :

3z 3z 1 X(z) = 2 = 2z 5z + 2 2 5z 1 + 2z 2 2 1 1 1 = 1 1 2z 1 = 1 1 z 1 1 2z 1 2z 2 Now n 1 x[n] = u[n] + 2n u[1 n] 2

x[n]

Check Yourself

Find the inverse transform of X(z) = 3z 2z 2 5z + 2

given that the ROC includes the unit circle.

x[n]

Solving Dierence Equations with Z Transforms

Start with dierence equation: 1 y[n] y[n 1] = [n] 2 Take the Z transform of this equation: 1 Y (z) z 1 Y (z) = 1 2 Solve for Y (z): 1 Y (z) = 1 1 z 1 2 Take the inverse Z transform (by recognizing the form of the trans form): n 1 u[n] y[n] = 2

Inverse Z transform

The inverse Z transform is dened by an integral that is not partic ularly easy to solve. Formally, x[n] = 1 X(z)z n1 ds 2j C were C represents a closed contour that circles the origin by running in a counterclockwise direction through the region of convergence. This integral is not generally easy to compute. This equation can be useful to prove theorems. There are better ways (e.g., partial fractions) to compute inverse transforms for the kinds of systems that we frequently encounter.

Properties of Z Transforms

The use of Z Transforms to solve dierential equations depends on several important properties. Property Linearity Delay Multiply by n Convolve in n

x[n] ax1 [n] + bx2 [n] x[n 1] nx[n] x1 [m]x2 [n m]

X(z) z 1 X(z) z dX(z)

dz

ROC

aX1 (z) + bX2 (z) (R1 R2 ) R R (R1 R2 )

m=

X1 (z)X2 (z)

Check Yourself

Find the inverse transform of Y (z) = 2

z z1

|z| > 1.

Check Yourself

Find the inverse transform of Y (z) = z z1 2 ; |z| > 1.

y[n] corresponds to unit-sample response of the right-sided system 2 2 2 Y 1 1 z = = = 1 X z1 1R

1z
= 1 + R + R2 + R3 + 1 + R + R2 + R3 + 1 1 R R2 R3 1 R R2 R3 R R R2 R3 R4 R2 R3 R2 R3 R3 R4 R4 R5 R5 R6 Y (n + 1)Rn = 1 + 2R + 3R2 + 4R3 + = X
n=0

y[n] = h[n] = (n + 1)u[n]

Check Yourself

Table lookup method. 2 z Y (z) = z1 z

z 1

y[n] =?
u[n]

Properties of Z Transforms

The use of Z Transforms to solve dierential equations depends on several important properties. Property Linearity Delay Multiply by n Convolve in n

x[n] ax1 [n] + bx2 [n] x[n 1] nx[n] x1 [m]x2 [n m]

X(z) z 1 X(z) z dX(z)

dz

ROC

aX1 (z) + bX2 (z) (R1 R2 ) R R (R1 R2 )

m=

X1 (z)X2 (z)

Check Yourself

Table lookup method. Y (z) = 2

z z1 z z 1
2
z 1 d =z z dz z 1 z1 2 z d z = z z dz z 1 z1

y[n] =?
u[n] nu[n]
(n + 1)u[n + 1] = (n + 1)u[n]

Concept Map: Discrete-Time Systems

Relations among representations. Block Diagram

X +
Delay

Delay R

System Functional
Y

+
Delay

Y 1 = X 1 R R2

Unit-Sample Response
index shift

h[n] : 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, . . .

Dierence Equation

y[n] = x[n] + y[n1] + y[n2]

System Function
Y (z) z2 H(z) = = X(z) 1 z z2

MIT OpenCourseWare http://ocw.mit.edu

6.003 Signals and Systems

Spring 2010

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.