EECE 301 Signals & Systems Prof.

Mark Fowler
Note Set #34
D-T Systems: Z-Transform Solving Difference Eqs. & Transfer Func. Reading Assignment: Sections 7.4 7.5 of Kamen and Heck

1/16

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

Ch. 3: CT Fourier Signal Models

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

Ch. 5: CT Fourier System Models

Frequency Response Based on Fourier Transform

Ch. 6 & 8: Laplace Models for CT Signals & Systems

Transfer Function

New System Model

New System Model

Ch. 2 Diff Eqs

C-T System Model Differential Equations D-T Signal Model Difference Equations Zero-State Response Zero-Input Response Characteristic Eq.

Ch. 2 Convolution
C-T System Model Convolution Integral D-T System Model Convolution Sum

New System Model

New Signal Model Powerful Analysis Tool

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

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

Transfer Function

New System Model

New System Model 2/16

ZT For Difference Eqs.

Given a difference equation that models a D-T system we may want to solve it: -with ICs -with ICs of zero
Apply ZT to the Difference Equation Use the Transfer Function Approach

Note the ideas here are very much like what we did with the Laplace Transform for CT systems.

Well consider the ZT/Difference Eq. approach first

3/16

Solving a First-order Difference Equation using the ZT

Given : y[n ] + ay[n 1] = bx[n ] IC = y[ 1] x[n ] for n = 0, 1, 2 , ...

Take ZT of differential equation: Z {y[n ] + ay[n 1]} = Z {bx[n ]} Use Linearity
of ZT

Solve for: y[n] for n = 0, 1, 2,

Z {y[n ]}+ aZ {y[n 1]} = bZ {x[n ]}

Y(z) X(z)
Need Right-Shift Property but which one???

Because of the non-zero IC we need to use the non-causal form:

Z {y[n 1]} = z 1Y ( z ) + y[1]

4/16

Using these results gives: Y ( z ) + a z 1Y ( z ) + y[ 1] = bX ( z ) Which is an algebraic equation that can be solved for Y(z):

b ay[ 1] Y ( z) = X ( z) + 1 1 1 + az 1 + az

Not the best form for doing Inverse ZT we want things in terms of z not z-1

Multiply each term by z/z

z bz Y ( z ) = ay[ 1] + X ( z) z+a z+a

On ZT Table

bz Part due to input signal H ( z) = modified by Transfer Function z+a

y[n ] = ay[ 1]( a ) n u[n ] + Z 1{H ( z ) X ( z )}
If |a| < 1 this dies out as n , its an IC-driven transient If the ICs are zero, this is all we have!!!
5/16

Ex.: Solving a Difference Equation using ZT: 1st-Order System w/ Step Input z For x[n ] = u[n ] X ( z ) = z 1 Then using our general results we just derived we get:

Y ( z) =

ay[1]z bz z + z+a z + a z 1

For now assume that a -1 so we dont have a repeated root. Then doing Partial Fraction Expansion we get (and we have to do the PFE by hand because we dont know a but it is not that hard!!!)

ay[ 1]z ( aab1 )z ( ab 1 )z Y ( z) = + + + + z+a z+a z 1

b y[n ] = ay[ 1]( a ) + a ( a )n + (1) n a +1
n

Now using ZT Table we get:

n = 0, 1, 2,...

IC-Driven Transient: decays if system is stable

Input-Driven Output 2 Terms: 1st term decays (Transient) 2nd term persists (Steady State)

6/16

Solving a Second-order Difference Equation using the ZT The Given Difference Equation:

y[n] + a1 y[n 1] + a2 y[n 2] = b0 x[n] + b1 x[n 1]

Assume that the input is causal Assume you are given ICs: y[-1] & y[-2] Find the system response y[n] for n = 0, 1, 2, 3, Take the ZT using the non-causal right-shift property:

Y ( z ) + a1 (z 1Y ( z ) + y[ 1]) + a2 (z 2Y ( z ) + z 1 y[ 1] + y[ 2]) = b0 X ( z ) + b1 z 1 X ( z )

a1 y[ 1] a2 y[ 1]z 1 a2 y[ 2] b0 + b1 z 1 + Y ( z) = X ( z) 1 2 1 2 1 + a1 z + a2 z 1 + a1 z + a2 z
Due to ICs decays if system is stable

H(z)

Due to input will have transient part and steady-state part

7/16

Lets take a look at the IC-Driven transient part:

a1 y[ 1] a2 y[ 1]z 1 a2 y[ 2] A Bz 1 Yzi ( z ) = = 1 2 1 + a1 z + a2 z 1 + a1 z 1 + a2 z 2
Multiply top and bottom by z2:

Az 2 + Bz Yzi ( z ) = 2 z + a1 z + a2

Now to do an inverse ZT on this requires a bit of trickery Take the bottom two entries on the ZT table and form a linear combination:

C1a n cos(o n )u[n ] + C2 a n sin(o n )u[n ]

C1 z 2 + a (C2 sin(o ) C1 cos(o ) )z z 2 2a cos(o ) z + a 2

a = a2 C1 = A

a1 0 = cos 2 a2 cos(0 ) B C1 C2 = sin(0 ) a sin(0 )

1

Compare & Identify

8/16

Finally, by a trig ID we know that

C1a n cos(o n )u[n ] + C2a n sin(o n )u[n ] = Ca n cos(o n + )u[n ]

So all of this machinery leads to the insight that the IC-Driven transient of a second-order system will look like this:

y zi [n ] = Ca n cos(o n + )u[n ]
where: 1. The frequency 0 and exponential a are set by the Characteristic Eq. 2. The amplitude C and the phase are set by the ICs
Note: If |a2| < 1 then we get a decaying response!!
9/16

a = a2

a1 0 = cos 2 a2
1

Solving a Nth-order Difference Equation using the ZT

y[n] + ai y[n i ] = bi x[n 1]

i =1 i =0

Contains x[n], x[n-1], If this system is causal, we wont have x[n+1], x[n+2], etc. here

A( z ) = z N + a1 z N 1 + ... + a N 1 z + a N B( z ) = b0 z N + b1 z N 1 + ... + bM z N M C ( z ) = depends on the IC' s

Transforming gives:

C ( z ) B( z ) Y ( z) = + X ( z) A( z ) A( z )
H(z) transfer function Transient part due to ICs

Transient and steady state part due to input

10/16

Discrete-Time System Relationships

Time Domain Z / Freq Domain
Pole/Zero Pole/Zero Diagram Diagram Roots Difference Difference Equation Equation
y[ k ] + an 1 y[k 1] + bn f [k ] + bn 1 f [k 1] +

ZT (Theory) Inspect (Practice)

Transfer Transfer Function Function

H ( z) =

bn z n + bn 1z n 1 + + b1 z + b0 z n + an 1 z n 1 + + a1z + a0

+ a1 y[k (n 1)] + a0 y[k n] = + b1 f [k (n 1)] + b0 f [k n]

Unit Circle ZT DTFT Frequency Frequency Response Response

Impulse Impulse Response Response h[k]

H ( ) = H ( z ) | z = e j
11/16

Example System Relationships

Time Domain
Difference Difference Equation Equation ZT (Theory) Inspect (Practice)

Z / Freq Domain
Transfer Transfer Function Function

y (n) y (n 1) = x(n) y (n) = x(n) + y (n 1)

Recursion Form

Input-Output Form

Y (z) 1 z = X (z)
1

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

12/16

Example System (cont.)

Z / Freq Domain

z = e j z 1 = e j
On Unit Circle

H ( z) = 1 z 1
z 1 = e j

Transfer Transfer Function Function Unit Circle Frequency Frequency Response Response

H ( ) = 1 e j ( , ]

13/16

Example System (cont.)

Plotting Transfer Function
= 1 e j 1 [ cos() j sin()] = [1 cos()] + j sin()
H ( ) = H () =
Eulers Equation

Group into Real & Imag

[1 cos()] + [ sin()]
2

sin() H () = tan 1 1 cos()

Standard Eqs for Mag. & Angle

14/16

Example System (cont.)

Unit Circle

Z/Freq Domain
Pole/Zero Pole/Zero Diagram Diagram

Im{z}

Re{z} Roots Transfer Transfer Function Function

z H ( z) = = 1 1 z z

Zero Num. = 0 When z = 0 Pole Den. = 0 When z = If < 1: Inside UC If > 1: Outside UC
15/16

Example System (cont.)

Time Domain
Inv. ZT Table

Z / Freq Domain

H ( z) = 1 z 1
Transfer Transfer Function Function

h(n) =

ZT Unit Circle

Impulse Impulse Response Response

DTFT

Frequency Frequency Response Response

Inv. DTFT h( n) = d 1 e j Use Table if Possible

H ( ) = 1 e j ( , ]
16/16