Some Relevant Questions

▶ What are Discrete Time Signals?

▶ How do they arise?
▶ What do we want to do with them?
▶ What are the potential applications?
▶ Why discrete time signals and processing as opposed to continuous
(analog) time signal processing?
Discrete Time Signals

Discrete Time signal: Sequence of numbers denoted by {x[n]}. If the

numbers (values) are real, we have a real sequence. If complex, we have
a complex sequence.
Examples:
1. x[n] = 3n. A real sequence. x[0] = 0, x[1] = 3, .....
π √
2. x[n] = e j 4 n . A complex sequence, (j = −1.)
Origin of these Sequences

Usually obtained by sampling a continuous time sequence.

x[n] = xc (nT ), where T is the sampling interval.
How should we choose T ?
▶ T large preferred. Have to store smaller number of samples. For
real-time applications, provides for more time to carry out the
computations needed.
▶ Problem: could loose information. T small allows for a better
approximation of the continuous time signals by its samples.
1
Goal: Choose T as large as possible or the sampling frequency Fs = T as
small as possible subject to fidelity constraints.
Sampling Theorem: If the highest frequency in the real continuous time
signal xc (t) is Fmax = B Hz, i.e. |Xc (jΩ)| = 0, |Ω| > 2πB rads/sec
(band limited),and the signal is sampled at a rate
Fs = T1 ≥ 2Fmax = 2B Hz, then xc (t) can be exactly recovered from the
sample values x[n] = xc (nT ).
Goals of discrete time signal processing

▶ Analysis: Analyze signals for structure. Concepts of DFT, FFT,

become handy
▶ Processing: Process signals to achieve an objective. Filtering. This
leads to filter design (FIR, IIR).
Applications

▶ Speech and Audio

▶ Images and Video
▶ Digital Communications
▶ Biomedical applications
▶ Radars and Sonars
▶ ...
LTE Wireless Transmitter(Orthogonal Frequency Division
Multiplex (OFDM))

OFDM Transmitter

X0 x0

Serial Parallel
Input Add
to IFFT to
Symbols Parallel CP
Serial

XN-1 xN-1

RF Section DAC Windowing

LTE Wireless Receiver

OFDM Receiver

x0 X0
ADC Parallel
and Serial to to Serial Output
FFT
Remove Parallel and Symbols
CP Decoder

xN-1 XN-1
Digital versus Analog
Benefits of Discrete Time Processing

▶ Analog components tend to drift and have to be more accurately

chosen. Digital circuits are more robust.
▶ Programmable - Changing filters is easy
▶ Processor can be shared
▶ Cascade filters and there is no loading problem
▶ Low-frequency systems: large inductors and capacitors
Analog systems: Suitable for large bandwidth systems, low power systems
Discrete Time Processing versus Digital Signal Processing
▶ Discrete Time Processing: Signal sampled (Time discrete) and
samples are represented with infinite precision
▶ Digital Signal Processing: Signal sampled (Time discrete) and
samples are represented with finite precision, e.g. 8 or 12 bits.
 Discrete-Time Convolution
• The output of a linear shift-invariant system is
the convolution of the input signal and the
unit sample response:
Discrete Time Sequences

Sequence of numbers denoted by

x = {x[n]}

where x[n] is the nth number in the sequence, and n is an integer.

Sometimes we will just use x[n] and use context to distinguish between
the sequence and the value at time n.

Often these sequences arise by sampling a continuous time signal

x[n] = xa (nT )

and so we refer to n as the time variable.

Typical Sequences
1. Unit Sample or Delta function (Kronecker) :

1 n=0
x[n] = δ[n] =
0 n ̸= 0

2. Delayed delta function :


1 n = n0
x[n] = δ[n − n0 ] =
0 n ̸= n0

3. Step function : 
1 n≥0
x[n] = u[n] =
0 n<0

4. Real Exponential:
x[n] = Aαn u[n].

5. Sines, Cosines, Complex Exponentials: x[n] = A cos(ω0 n + ϕ)

Plots of Typical Sequences

1
Unit sample
... ...

0 n
(a)

Unit step
1
... ...

0 n
(b)

Real exponential
...

...
0 n
(c)

Sinusoidal

...

0 n
...
General Representation

In general, any signal can be written as a weighted sum of delayed delta

functions:
∞
X
x[n] = x[k]δ[n − k]
k=−∞
= . . . + x[−1]δ[n + 1] + x[0]δ[n] + x[1]δ[n − 1] + . . .
Example of a Sequence
Consider a sequence


 4, n=0
2, n=2

x[n] =

 3, n=5
0, otherwise


4
x[n]

0
-4 -2 0 2 4 6 8 10
n

This sequence can be considered a sum of 3 sequences, one for each

nonzero value
x[n] = 4δ[n] + 2δ[n − 2] + 3δ[n − 5]
 1
Unit sample
Another Example
... ...


1,
0 n≥0 n
x[n] =
0, otherwise
(a)

Unit step
1
... ...

Oppenheim, Schafer, and Buck ©1999-2000 Prentice Hall, Inc.

0 n
(b)
This sequence can be considered a sum of infinite sequences, one for
each nonzero value
∞
X
x[n] = δ[n − k].
k=0
P∞ Real exponential
In general x[n] = k=−∞... x[k]δ[n − k].
Typical Sequences

1. Delta function (unit sample sequence) :


1 n=0
x[n] = δ[n] =
0 n≠ 0


1 n = n0
2. Delayed delta function : x[n] = δ[n − n0 ] =
0 n ̸= n0

1 n≥0
3. Step function : x[n] = u[n] =
0 n<0
4. Sines, Cosines, Exponentials: x[n] = A cos(ω0 n + ϕ)
In general, any signal can be written as a weighted sum of delayed delta
functions:
X∞
x[n] = x[k]δ[n − k].
k=−∞
Discrete Time Systems

The input-output relation is given by y [n] = T {x[n]}

Too general for characterization and design purposes and so constraints
are needed.
Special Class of Systems

Memoryless Systems: Output at time n depends only on x[n], i.e input

at time n. No memory
Linear Systems: T {ax1 [n] + bx2 [n]} = aT {x1 [n]} + bT {x2 [n]}. Additivity
and Scaling property
Time-Invariant Systems: If y [n] = T {x[n]}, then
T {x[n − n0 ]} = y [n − n0 ]. Delayed input leads to delayed
output
Causality: Output y [n] at time n = n0 depends only on the input
sequence values for n ≤ n0 . Output depends only on the
present and past
Stable Systems: If |x[n] ≤ Bx < ∞ ∀n, then |y [n]| ≤ By < ∞ ∀n. Must
be true for all bounded inputs. Such systems are
Bounded-Input Bounded-Output (BIBO) stable
Examples

System 1: y [n] = x 2 [n], Memoryless: yes, Linear: no, Time-Invariant:

yes, Causal: yes, Stable: yes
System 2: y [n] = A cos(ω0 n + ϕ)x[n], Memoryless: yes, Linear: yes,
Time-Invariant: No, Causal: yes, Stable: yes
System 3: y [n] = 12 (x[n] + x[n − 1]), Memoryless: no, Linear: yes,
Time-Invariant: yes, Causal: yes, Stable: yes
System 4: y [n] = 13 (x[n] + x[n − 1] + x[n + 1]), Memoryless: no,
Linear: yes, Time-Invariant: yes, Causal: no, Stable: yes
Linear Time-Invariant (LTI) Systems
Reason: Simple characterization and design. Response to a unit sample
sequence (delta function) is sufficient to define the system
Suppose h[n] = T {δ[n]}, then since system is time-invariant
T {δ[n − k]} = h[n − k]. h[n] is the impulse response of the system.
For LTI systems
∞
X
y [n] = T {x[n]} = T { x[k]δ[n − k]}
k=−∞
∞
X
= x[k]T {δ[n − k]} (Linearity)
k=−∞
X∞
= x[k]h[n − k] (Time-Invariance)
k=−∞
= x[n] ∗ h[n] (Convolution)
If we expand the sum
y [n] = ... + x[−1]h[n + 1] + x[0]h[n] + x[1]h[n − 1] + x[2]h[n − 2] + ...
Suggests a procedure for carrying out convolution
Alternate Approach to Convolution

P∞
y [n] = k=−∞ x[k]h[n − k]

1: h1 [k] = h[−k] (Flip sequence about origin)

2: Construct h[n − k] for particular n. Amounts to shifting
h1 [k] to the right by n for n > 0, and to the left by |n| for
n < 0.
3: Multiply x[k] and h[n − k] term by term to obtain
x[k]h[n − k] for each value of k.
4: Sum all thePvalues to get y [n],
∞
i.e.y [n] = k=−∞ x[k]h[n − k]
5: Repeat above steps (2-4) for all n.
Properties of Convolution

Commutative: x[n] ∗ h[n] = h[n] ∗ x[n]. System and input can be

interchanged
Distributive: x[n] ∗ (h1 [n] + h2 [n]) = x[n] ∗ h1 [n] + x[n] ∗ h2 [n]. Leads to
the parallel form
Associative: (x[n] ∗ h1 [n]) ∗ h2 [n] = x[n] ∗ (h1 [n] ∗ h2 [n]). leads to
cascade forms
Also by the commutative property,
x[n] ∗ (h1 [n] ∗ h2 [n]) = x[n] ∗ (h2 [n] ∗ h1 [n]). This allows reordering of
systems
Distributive Property
x[n] ∗ (h1 [n] + h2 [n]) = x[n] ∗ h1 [n] + x[n] ∗ h2 [n].
Commutative and Associative Property
(x[n] ∗ h1 [n]) ∗ h2 [n] = x[n] ∗ (h1 [n] ∗ h2 [n]).
Causality and Stability of LTI Systems

LTI systems defined by their impulse response h[n].

P∞
Output for any input can be computed as y [n] = k=−∞ x[k]h[n − k]
Causality and Stability can be inferred from h[n].
System 1: h[n] = 2δ[n] + 3δ[n − 4] + 3δ[n − 5]
System 2: h[n] = δ[n + 2] + 2δ[n] + 3δ[n − 4] + 3δ[n − 5]
System 3: h[n] = u[n].
Causality LTI Systems

Causality: h[n] = 0, ∀n < 0.

Proof: P
∞ Pn P∞
y [n] = k=−∞ x[k]h[n−k] = k=−∞ x[k]h[n−k]+ k=n+1 x[k]h[n−k]
First term depends on the past values of the input. The second term
depends on the future values and should be zero for causal systems. This
requires h[n − k] = 0, k = n + 1, n + 2, ... or h[n] = 0, n < 0.
For a causal system
n
X ∞
X
y [n] = x[k]h[n − k] = h[k]x[n − k]
k=−∞ k=0

In the examples mentioned: System 1 and 3 are causal, System 2 is

non-causal.
Stability for LTI Systems

Based on h[n], how do you determine stability?

Main Result: A LTI system is BIBO stable

P∞ if and only if the impulse
response is absolutely summable, i.e k=−∞ |h[k]| < ∞.

In the examples mentioned: System 1 and 2 are stable, System 2 is

unstable.
Linear Constant-Coefficient Difference Equations

Even general
P∞ LTI systems are not practical. Why?
y [n] = k=0 h[k]x[n − k] and requires an infinite sum.
This leads to LTI systems described by difference equations
N
X M
X
ak y [n − k] = bm x[n − m]
k=0 m=0

or alternatively
N M
X ak X bm
y [n] = − y [n − k] + x[n − m]
a0 a
m=0 0
k=1

Can set a0 = 1 without loss of generality (wlog). ak are referred to as the

feedback coefficients and bm the feedforward coefficients.
No infinite sums and so viable. Design reduces to choosing N, M, ak ’s
and bm ’s.
FIR and IIR Filters

Finite Impulse Response (FIR) filters:

M
X
y [n] = bm x[n − m]
m=0

ImpulsePresponse of a FIR filter: Compare with convolution

∞
y [n] = k=0 h[k]x[n − k].
h[n] = bn , 0 ≤ n ≤ M, and zero otherwise.
Can be implemented using simple memory, multipliers and adders.
Is it stable?Yes, as long as the bm are bounded.
IIR filters
N
X M
X
y [n] = − ak y [n − k] + bm x[n − m]
k=1 m=0

Because of the feedback terms, impulse response is of infinite duration.

Stability depends on the feedback coefficients.
Simple IIR Example

y [n] = ay [n − 1] + x[n]
Assume system at rest, i.e. y [−1] = 0. (Clear memory)
Setting x[n] = δ[n] leads to the output y [n] = h[n].
Can show
h[n] = an u[n]
Stable? Based on absolute summability criteria, Stable if |a| < 1.
Discrete Time Fourier Transform

∞
X
X (e jω ) = x[n]e −jωn
n=−∞
Z π
1
x[n] = X (e jω )e jωn dω
2π −π

Will use the notation X (e jω ) = F(x[n]). X (e jω ) is periodic with

periodicity 2π, X (e jω ) = X (e j(ω+2π) ).
Some Observations
▶ ω is referred to as the normalized frequency.
▶ Definition consistent with the definition of the continuous time
Fourier transform and related through the sampling theorem.
▶ x[n] can be viewed as the Fourier series coefficients.
Fourier Series Connection

DTFT
∞
X
X (e jω ) = x[n]e −jωn
n=−∞
Z π
1
x[n] = X (e jω )e jωn dω
2π −π

Fourier Series
∞
X 2π
xc (t) = xc (t + T0 ) = ck e jkΩ0 t , where Ω0 =
T0
k=−∞
Z T0
1
ck = xc (t)e −jkΩ0 t dt
T0 0

Connection X (e jω ) ↔ xc (t) and x[n] ↔ ck .

At a variable level we have t ↔ ω, T0 ↔ 2π, and Ω0 ↔ 1.
Magnitude and Phase

jω
X (e jω ) = |X (e jω )|e j∡X (e )
= XR (e jω ) + jXI (e jω )
jω
XR2 (e jω ) + XI2 (e jω ) and ∡X (e jω ) = arctan XXRI (e )
p
|X (e jω )| = (e jω )

▶ Magnitude is usually plotted in db scale, i.e. 20 log10 |X (e jω )|.

▶ Phase: there are two popular options
▶ ARG(X (e jω )) is ∡X (e jω ) limited to the range [−π, π].
▶ arg(X (e jω )) is ∡X (e jω ) computed as a continuous function of
ω (unwrapped phase).
Examples

Delta function:
∞
X
x[n] = δ[n] ↔ X (e jω ) = δ[n]e −jωn = 1.
n=−∞

Delayed Delta function:

∞
X
x[n] = δ[n−n0 ] ↔ X (e jω ) = δ[n−n0 ]e −jωn = e −jωn0 .
n=−∞

Exponential sequence: x[n] = an u[n], |a| < 1, has DTFT

∞ ∞
X X 1
X (e jω ) = an u[n]e −jωn = (ae −jω )n =
n=−∞ n=0
1 − ae −jω
Time Shifting

Time Shifting: x[n − nd ] ↔ e −jωnd X (e jω )

Proof:
∞
X
F(x[n − nd ]) = x[n − nd ]e −jωn
n=−∞
X∞
= x[m]e −jω(m+nd ) (Change of variables m = n − nd )
m=−∞
∞
X
= e −jωnd x[m]e −jωm = e −jωnd X (e jω )
m=−∞
Modulation

Modulation: e jω0 n x[n] ↔ X (e j(ω−ω0 ) )

Proof:
∞
X
F(e jω0 n x[n]) = e jω0 n x[n]e −jωn
n=−∞
X∞
= x[n]e −j(ω−ω0 )n
n=−∞

= X (e j(ω−ω0 ) )
Convolution

x[n] ∗ y [n] ↔ X (e jω )Y (e jω )
Proof:
∞
X
F(x[n] ∗ y [n]) = F( x[k]y [n − k])
k=−∞
∞
X
= x[k]F(y [n − k]) (Linearity )
k−∞
X∞
= x[k]e −jωk Y (e jω ) (Time Shifting )
k−∞

= X (e jω )Y (e jω )
LTI Systems and Convolutions

y [n] = h[n] ∗ x[n] ↔ Y (e jω ) = H(e jω )X (e jω )

h[n] ↔ H(e jω ) and H(e jω ) is the frequency response (transfer function)
of the LTI system.
LTI systems and complex exponential inputs: Consider input x[n] = e jω0 n .
∞
X ∞
X
y [n] = h[k]x[n − k] = h[k]e jω0 (n−k)
k=−∞ k=−∞
∞
X
= e jω0 n h[k]e −jω0 k = e jω0 n H(e jω0 )
k=−∞

Complex exponential input leads to a scaled complex exponential at the

output with the scaling determined by the Transfer function. They are
referred as eigenfunctions of LTI systems.
jωk n
P
Input: x[n]
P = kjωckk e jωk n. Output: By the linearity property
y [n] = k ck H(e )e
Parseval’s theorem

∞ Z π
X 1
|x[n]|2 = |X (e jω )|2 dω
n=−∞
2π −π


1, −ωc ≤ ω ≤ ωc R ωc sin ωc n
X (e jω ) = and x[n] = 1
2π −ωc
e −jωn dω = πn
0, Otherwise
From Parseval’s Theorem
∞ ∞  2 Z ωc
X X sin ωc n 1 ωc
|x[n]|2 = = dω =
n=−∞ n=−∞
πn 2π −ωc π
Property 1: x ∗ [n] ↔ X ∗ (e −jω ).

x[n] = xR [n] + jxI [n] ↔ X (e jω ) = XR (e jω ) + jXI (e jω ), then

x ∗ [n] = xR [n] − jxI [n] ↔ X ∗ (e −jω ) = XR (e −jω ) − jXI (e −jω ).
Note: xR [n] and XR (e jω ) are not Fourier transform pairs, i.e.
xR [n] ̸ ↔XR (e jω ).
Proof:
∞
X ∞
X ∞
X
F(x ∗ [n]) = x ∗ [n]e −jωn = x ∗ [n](e jωn )∗ = (x[n]e jωn )∗
n=−∞ n=−∞ n=−∞
∞
!∗ ∞
!∗
X X
jωn −j(−ω)n
= (x[n]e ) = (x[n]e )
n=−∞ n=−∞

∗
= X (e −jω ) = X ∗ (e −jω )
 ∗ ∗
jω
(e −jω )
Property 2: Re(x[n]) = xR [n] = x[n]+x
2
[n]
↔ Xc (e jω ) = X (e )+X2 .
jω ∗ −jω
Xc (e ) = Xc (e ) (Conjugate-Symmetric part of X (e jω ).)
∗ jω ∗ −jω
Property 3: jIm(x[n]) = jxI [n] = x[n]−x
2
[n]
↔ Xo (e jω ) = X (e )−X
2
(e )
.
Xo (e jω ) = −Xo∗ (e −jω ) (Conjugate-Antisymmetric part of X (e jω ).)
Real Sequences

For a real sequence x[n] = x ∗ [n]. Hence X (e jω ) = X ∗ (e −jω ).

Implications:
|X (e jω )| = |X ∗ (e −jω )| = |X (e −jω )|. The magnitude of the DTFT is an
even function. Sufficient to plot [0, π].

∡X (e jω ) = ∡X ∗ (e −jω ) = −∡X (e −jω )|. The phase of the DTFT is an

odd function. Sufficient to plot [0, π].

X (e jω ) = X ∗ (e −jω ) → XR (e jω ) + jXI (e jω ) = XR (e −jω ) − jXI (e −jω ).

Hence XR (e jω ) = XR (e −jω ) and XI (e jω ) = −XI (e −jω ).
Convergence of the Fourier Transform

P∞
X (e jω ) = n=−∞ x[n]e −jωn . Infinite sum and so may not exist for a
given x[n].
PM
XM (e jω ) = n=−M x[n]e −jωn −−−−→
M→∞ X (e jω )
Complicated because it involves a sequence of functions. Behavior may
vary with ω.
Pointwise Convergence: limM→∞ XM (e jω ) = X (e jω ) ∀ω.
Uniform Convergence: {XM (e jω )} converges uniformly to X (e jω ) if given
any ϵ > 0, there exists a natural number N = N(ϵ) such that

|XM (e jω ) − X (e jω )| < ϵ, for every M > N and ∀ω.

Uniform convergence implies pointwise convergence but not the other

way around.
The limit of a sequence of continuous functions converging uniformly is
also continuous.
Absolute Summability and Convergence

If the sequence is absolutely summable, the Fourier transform exists.

∞
X ∞
X ∞
X
|X (e jω )| = | x[n]e −jωn | ≤ |x[n]e −jωn | ≤ |x[n]| < ∞
n=−∞ n=−∞ n=−∞

Can show: Converges uniformly to a continuous function.

For stable LTI systems, the Fourier transform always exists.
Reason: Stable LTI systems have an impulse response that is absolutely
summable.
Convergence Issues with Low Pass Filters


1, −ωc ≤ ω ≤ ωc R ωc sin ωc n
jω
H(e ) = and h[n] = 1
2π −ωc
e −jωn dω = πn
0, Otherwise

Challenges:
▶ Non-causal and infinite in duration
▶ HM (e jω ) = M −jωn
P
n=−M h[n]e does not converge uniformly. Note
jω
P M −jωn
that HM (e ) = n=−M h[n]e is a continuous function of ω
but the ideal lowpass filter is not
▶ The ideal impulse response is not absolutely summable

Consequence: Gibbs Phenomenon

Gibbs Phenomenon
Continuous Time Fourier Transform

Laplace Transform Z ∞
Xc (s) = xc (t)e −st dt
−∞

Continuous Time Fourier Transform

Z ∞
Xc (jΩ) = Xc (s)|s=jΩ = xc (t)e −jΩt dt
−∞
Z ∞
1
xc (t) = Xc (jΩ)e jΩt dΩ
2π −∞

Ω = 2πF , where Ω is in radians per sec and F is in Hz or cycles per sec.

Fourier Series

∞
X 2π
xp (t) = xp (t + T0 ) = ck e jkΩ0 t , where Ω0 =
T0
k=−∞
T0
Z T0 Z
1 −jkΩ0 t 1 2
ck = xp (t)e dt = xp (t)e −jkΩ0 t dt
T0 0 T0 T
− 20

What is the Fourier transform of a periodic function? For this we will use
the following Fourier transform pair

e jΩ0 t ↔ 2πδ(Ω − Ω0 )

Fourier transform of a periodic function by linearity and Fourier series

expansion is
X∞
Xp (jΩ) = 2π ck δ(Ω − kΩ0 )
k=−∞
Some properties of Dirac Delta functions

Rϵ
1. −ϵ
δ(t)dt = 1, ϵ > 0

2. f (t)δ(t − t0 ) = f (t0 )δ(t − t0 )

R∞
3. −∞
f (t)δ(t − t0 )dt = f (t0 )

4. f (t) ∗ δ(t − t0 ) = f (t − t0 )
Ideal Continuous to Discrete Converter
 P
Sampling with a Pulse Train s(t) = n δ(t − nT )
Mathematical View of Sampling

∞
X
s(t) = δ(t − nT )
n=−∞

is a periodic function with periodicity T . ThePfundamental frequency

∞
Ωs = 2πT and the Fourier series expansion is k=−∞ ck e
jkΩs t
with the
Fourier series coefficients given by
Z T Z T ∞
1 2
−jkΩs t 1 2 X 1
ck = s(t)e dt = δ(t − nT )e −jkΩs t dt =
T − T2 T − T2 n=−∞
T

Fourier Transform of s(t) is given by

∞
2π X
S(jΩ) = δ(Ω − kΩs )
T
k=−∞
Sampled function (Time Domain View)

∞
X
xs (t) = xc (t)s(t) = xc (t) δ(t − nT )
n=−∞
∞
X ∞
X
= xc (t)δ(t − nT ) = xc (nT )δ(t − nT )
n=−∞ n=−∞
X∞
= x[n]δ(t − nT ), where x[n] = xc (nT )
n=−∞

There is a natural connection between the sampled function and the

sequence
xs (t) ↔ {x[n]}
Given xs (t), one can obtain the sequence {x[n]} by stripping away the
impulses. P∞
Given {x[n]}, one can obtain xs (t) as xs (t) = n=−∞ x[n]δ(t − nT ).
Sampled Function and Corresponding Sequence

∞
X
xs (t) = x[n]δ(t − nT )
n=−∞
Sampled Function (Frequency Domain View)

1
xs (t) = xc (t)s(t) ↔ Xs (jΩ) = [Xc (jΩ) ∗ S(jΩ)]
2π

∞
" !#
1 1 2π X
Xs (jΩ) = [Xc (jΩ) ∗ S(jΩ)] = Xc (jΩ) ∗ δ(Ω − kΩs )
2π 2π T
k=−∞
∞ ∞
1 X 1 X
= [Xc (jΩ) ∗ δ(Ω − kΩs )] = Xc (j(Ω − kΩs ))
T T
k=−∞ k=−∞
1
= {. . . Xc (j(Ω + Ωs )) + Xc (jΩ) + Xc (j(Ω − Ωs )) + . . .}
T
Recovery is possible if the shifted copies (images) do not overlap with
Xc (jΩ), i.e. no aliasing
Frequency Domain Representation of Sampling
Nyquist-Shannon Sampling Theorem

Let xc (t) be a bandlimited signal with

Xc (jΩ) = 0, for |Ω| ≥ ΩN .

Then xc (t) is uniquely determined by its samples

x[n] = xc (nT ), n = 0, ±1, ±2, . . . iff

2π
Ωs = ≥ 2ΩN .
T
2ΩN is referred to as the Nyquist rate.
Recovery

The reconstruction filter is an ideal continuous-time low pass filter

Reconstruction Filter

|Ω| ≤ Ω2s sin( πt


T T )
Hr (jΩ) = ↔ hr (t) = πt
0 otherwise T
Assuming no aliasing

Xc (jΩ) = Xs (jΩ)Hr (jΩ)

Time domain interpolation

∞
X ∞
X
xc (t) = xs (t) ∗ hr (t) = x[n]δ(t − nT ) ∗ hr (t) = x[n]hr (t − nT )
n=−∞ n=−∞

Exact reconstruction of the continuous-time signal from the samples.

Normalized Frequency (ω) versus Actual Frequencies (Ω)
∞
X ∞
X
xs (t) = x[n]δ(t − nT ) ↔ Xs (jΩ) = x[n]e −jΩTn
n=−∞ n=−∞

Now the DTFT of x[n] is defined as

∞
X
X (e jω ) = x[n]e −jωn
n=−∞

Because xs (t) ↔ {x[n]}, we want Xs (jΩ) ↔ X (e jω ) for consistency. This

leads to the following relationship between ω and Ω.

ω = ΩT .

So ω = 0 corresponds to Ω = 0, and ω = ±π corresponds to

Ω = ± Tπ = Ω2s .
Example: If the bandwidth of the signal is 4Khz, and we sample at
10Khz, there is no aliasing. When we mention ωo = π4 for the normalized
frequency variable, we are referring to a frequency of 5Khz/4 in the
continuous domain.
Digital Systems as substitute for Continuous time Systems
Example
Example Design

Problem: xc (t) is a continuous time signal with bandwidth 8 Khz and we

are interested in only retaining the frequencies from 1 KHz to 3 Khz.
Design a system to achieve this objective.

Continuous time system design: Design a bandpass filter with passband

from 1 to 3 Khz, i.e. 2π1000 rads/sec ≤ |Ω| ≤ 2π3000 rads/sec.

Digital system design: Sample the signal at Nyquist rate or above. Let us
choose a sampling rate of 20 Khz for the discussion. Design a digital
π
bandpass filter with pass band from 10 to 3π π 3π
10 , i.e. 10 ≤ |ω| ≤ 10 .
Z-Transform Definition

∞
X
X (z) = x[n]z −n
n=−∞

Will use the notation X (z) = Z(x[n]).

Definition not complete without specifying the Region Of Convergence
(ROC)
∞
X ∞
X
ROC : {z : |x[n]z −n | = |x[n]||z|−n < ∞}
n=−∞ n=−∞

Why Z -transforms?
▶ More general: Applicable to a larger class of signals
▶ Easier to manipulate: Complex variable theory can be useful
(Laurent series and associated results).
Z-Transform versus DTFT

X (e jω ) = X (z)|z=e jω
Since |e jω | = 1, DTFT is the z-transform evaluated on the unit circle.

For the Fourier Transform to be defined, ROC must include the unit
circle. Sequence must be absolutely summable.
Example 1: Right Sided Sequence
Consider x[n] = an u[n].
∞
X ∞
X ∞
X
X (z) = x[n]z −n = an u[n]z −n = an z −n
n=−∞ n=−∞ n=0
∞
X 1
= (az −1 )n = , |az −1 | < 1
n=0
1 − az −1

ROC : {z : |az −1 | < 1} = {z : |z| > |a|}

Example 2: Left Sided Sequence
Consider x[n] = −an u[−n − 1].
∞
X −1
X
X (z) = − an u[−n − 1]z −n = − an z −n
n=−∞ n=−∞
∞ ∞
X
−n n
X a−1 z
= − a z =− (a−1 z)n = − , |a−1 z| < 1
n=1 n=1
1 − a−1 z
az −1 a−1 z 1 1
= − −1 −1
= − −1 = , |a−1 z| < 1
az 1 − a z az − 1 1 − az −1
ROC : {z : |a−1 z| < 1} = {z : |z| < |a|}
General Sequences

1. Finite Duration Sequences

2. Right Sided Sequences
3. Left Sided Sequences
4. Two Sided Sequences
Finite Duration Sequences

1. Finite Duration Sequences: Sequences that are non-zero for

−∞ < N1 ≤ n ≤ N2 < ∞

∞
X N2
X
X (z) = x[n]z −n = x[n]z −n
n=−∞ n=N1

ROC: All z except possibly z = 0 or z = ∞ because z −n is infinity at

z = 0 for positive values of n and similarly for z = ∞ for negative values
of n.
Right Sided Sequences

2. Right Sided Sequences: Sequences that are zero for n < N1 < ∞
∞
X ∞
X
X (z) = x[n]z −n = x[n]z −n
n=−∞ n=N1

ROC: {z : |z| > rR } except possibly z = ∞

P∞
Reasoning: If n=N1 |x[n]||z|−n converges for |z| = r1 , it will converge
for all |z| > r1 since all terms with positive n get smaller

|x[n]| |x[n]|
|x[n]r1−n | = > = |x[n]r2−n |, r2 > r1 and n > 0
r1n r2n
Left Sided Sequences

3. Left Sided Sequences: Sequences that are zero for n > N2 > −∞

∞
X N2
X
X (z) = x[n]z −n = x[n]z −n
n=−∞ n=−∞

ROC: {z : |z| < rL } except possibly z = 0

PN2
Reasoning: If n=−∞ |x[n]||z|−n converges for |z| = r1 , it will converge
for all |z| < r1 since all terms with negative n get smaller
Two-Sided Sequences
4. Two Sided Sequences: Sequences that are neither left-sided or
right-sided, i.e. sequence defined over −∞ < n < ∞.
X∞
X (z) = x[n]z −n
n=−∞

ROC: {z : rR < |z| < rL } is an annular ring

P∞ P−1 P∞
Reasoning: If n=−∞ x[n]z −n = n=−∞ x[n]z −n + n=0 x[n]z −n , sum
of a left sided sequence and a right sided sequences. Both sequences
must converge, and the ROC is the intersection.
Summary of ROCs

1. Finite Duration Sequences. ROC: All z except possibly z = 0 or

z =∞
2. Right Sided Sequences. ROC: {z : |z| > rR } except possibly z = ∞
3. Left Sided Sequences. ROC: {z : |z| < rL } except possibly z = 0
4. Two Sided Sequences. ROC: {z : rR < |z| < rL } is an annular ring
Examples

Delta function:
∞
X
x[n] = δ[n] ↔ X (z) = δ[n]z −n = 1.
n=−∞

ROC: Entire z-plane

Delayed Delta function:
∞
X
x[n] = δ[n − n0 ] ↔ X (z) = δ[n − n0 ]z −n = z −n0 .
n=−∞

ROC: Entire z-plane except z = 0 or z = ∞.

Exponential sequence: x[n] = an u[n], has Z-transform
∞ ∞
X X 1
X (z) = an u[n]z −n = (az −1 )n = , |az −1 | < 1
n=−∞ n=0
1 − az −1

ROC: {z : |az −1 | < 1} = {z : |z| > |a|}

Property 2: Time Shifting

x[n] ↔ X (z), ROC: Rx = {z : rR < |z| < rL }

Time Shifting: x[n − n0 ] ↔ z −n0 X (z), ROC is Rx (except the addition or
deletion of 0 and ∞ for sequences that are not two-sided)
Proof:
∞
X
Z(x[n − n0 ]) = x[n − n0 ]z −n
n=−∞
X∞
= x[m]z −(m+n0 ) (Change of variables m = n − nd )
m=−∞
∞
X
−n0
= z x[m]z −m = z −n0 X (z)
m=−∞

Due to the presence of z −n0 , ROC is Rx (except the addition or deletion

of 0 and ∞ for sequences that are not two-sided)
Property 3: Multiplying by an Exponential Sequence

Modulation: z0n x[n] ↔ X ( zz0 ), with ROC {z : |z0 |rR < |z| < |z0 |rL }
Proof:
∞
X
Z(z0n x[n]) = z0n x[n]z −n
n=−∞
∞
X z −n
= x[n]( )
n=−∞
z0
z
= X( )
z0

ROC: {z : rR < | zz0 | < rL } = {z : |z0 |rR < |z| < |z0 |rL }
Property 9: Convolution

T
x[n] ∗ y [n] ↔ X (z)Y (z) and ROC contains Rx Ry
Proof:
∞
X
Z(x[n] ∗ y [n]) = Z( x[k]y [n − k])
k=−∞
∞
X
= x[k]Z(y [n − k]) (Linearity )
k−∞
X∞
= x[k]z −k Y (z) (Time Shifting )
k−∞
= X (z)Y (z)

T to be defined for the product to be defined. So

X (z) and Y (z) have
ROC contains Rx Ry
Property 5

x ∗ [n] ↔ X ∗ (z ∗ ) and ROC is Rx

Proof:
∞
X ∞
X ∞
X
∗ ∗ −n ∗ ∗ −n ∗
Z(x [n]) = x [n]z = x [n](z ) ) = (x[n](z ∗ )−n ))∗
n=−∞ n=−∞ n=−∞
∞
!∗
X
= x[n](z ∗ )−n
n=−∞
∗ ∗
= (X (z )) = X ∗ (z ∗ )

P∞ P∞
ROC: n=−∞ |x ∗ [n]||z −n | = n=−∞ |x[n]||z −n |, and so the ROC is Rx
Example

1 1 1
X (z) = + +
1 − az −1 1 − bz −1 1 − cz −1

What are the possible ROCs?

Potential ROCs
 1 1 1
X (z) = −1
+ −1
+
1 − az 1 − bz 1 − cz −1
Consider ROC: {z : a < |z| < b}

What is x[n]?

1 1 1
X (z) = −1
+ −1
+
1 − az 1 − bz 1 − cz −1
| {z } | {z } | {z }
|z|>a |z|<b |z|<c

x[n] = an u[n] − b n u[−n − 1] − c n u[−n − 1]

 1 1 1
X (z) = −1
+ −1
+
1 − az 1 − bz 1 − cz −1
Consider ROC: {z : |z| > c}

What is x[n]?

1 1 1
X (z) = −1
+ −1
+
1 − az 1 − bz 1 − cz −1
| {z } | {z } | {z }
|z|>a |z|>b |z|>c

x[n] = an u[n] + b n u[n] + c n u[n]

 3
2
sha1_base64="yLKHvIcA189JjTbKLpBB6PlU9lM=">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</latexit>
<latexit
sha1_base64="x7LI8avslKZphpOCdpTstl4TuwE=">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</latexit>
<latexit
sha1_base64="XyTu1L+1nXQb8vyLRDKl0KtxYCQ=">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</latexit>
<latexit
x[n]

• Imp
sha1_base64="yLKHvIcA189JjTbKLpBB6PlU9lM=">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</latexit>
<latexit
Y (z) =

=
H(z) =
Delay

N
x[n]

k=1
• B: Shifting
sha1_base64="UmYY2x+bTNuRQrM6IuFPXYl98cw=">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</latexit>
<latexit
• A: Linearity

Y (z)
X(z)
ak z
sha1_base64="1L+kaD1DKWgBDgxGDJu9Cews9yk=">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</latexit>
<latexit
=
(delay/advance)

= H(z)X(z)
h[n]

1
y[n] = x[n
sha1_base64="5fxVpM0qxq2i7QQhu7+NVQCRKYA=">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</latexit>
<latexit
sha1_base64="oAp+C75Zud3StPyS3AXKAUDbpfY=">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</latexit>
<latexit
sha1_base64="80eMCXZstDk6Mrbw4cmmMFjwD98=">AAAC2nicfVJNb9NAEF27BUr4aChHLqNGoFQokV0q0UulSlw4Fqlpgmw3Wm/Wzcq7a2t3jXBdHziAhLjyy7jxL/gJrBO3ognqSCu9mXnvzX7FOWfaeN5vx93YvHf/wdbDzqPHT55ud5/tnOmsUISOSMYzNYmxppxJOjLMcDrJFcUi5nQcp++a/vgTVZpl8tSUOY0EvpAsYQQbW5p2/3zsX+7BqyMIdSHOq5DJxJT1tJJwBIM2gzKQEVyeVwNZw93Mz4GEAaQ37DDsrHvz2wq+ZPf563Sv8W+S1ErnOseEVoPhGyHqOz1WqDeW/HoDreXEnnXa7XlDbxGwDvwW9FAbJ9Pur3CWkUJQaQjHWge+l5uowsowwmndCQtN7fQUX9DAQokF1VG1eJkaXtrKDJJM2SUNLKr/KiostC5FbJkCm7le7TXF//WCwiSHUcVkXhgqyXJQUnAwGTTPDDOmKDG8tAATxexegcyxwsTYz9Cxl+CvHnkdjPeH/sHQ9z8c9I732/vYQi/QLuojH71Fx+g9OkEjRJwz58r56nxzA/eL+939saS6Tqt5jm6F+/MvBivgew==</latexit>
<latexit
PM
k]

Y (z) +

PN
y[n]

r=0
X
X(z) =

r=0 br z
br z

k=1 ak z
=
Y (z) =
= a1

r
=z
1

z N + a1 z N 1 + · · · + aN
X

b0 z M + b1 z M 1 + · · · + bM
n= 1

k
1

• C: System Input/Output Description

1

X
X

l= 1

X(z)
n= 1

n= 1

X(z)
x[l]z
y[n]z
sha1_base64="zIB8y1eZS1jarJFq8bQm/Crj/GE=">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</latexit>
<latexit
x1 [n]z

n
n
x[n] = a1 x1 [n] + a2 x2 [n]

(l+k)

=
y[n] =
=

N
+ a2

sha1_base64="iVsY2HZwWgN0cHD1Nv8XZ6jxP6Q=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBahp5JIQY8FLx4rWFtIQ9lsN+3S3U3Y3Ygl9C948aAgXv1B3vw3btIctPXBwOO9GWbmhQln2rjut1PZ2Nza3qnu1vb2Dw6P6scnDzpOFaE9EvNYDUKsKWeS9gwznA4SRbEIOe2Hs5vc7z9SpVks7808oYHAE8kiRrDJpSdfBqN6w225BdA68UrSgBLdUf1rOI5JKqg0hGOtfc9NTJBhZRjhdFEbppommMzwhPqWSiyoDrLi1gW6sMoYRbGyJQ0q1N8TGRZaz0VoOwU2U73q5eJ/np+a6DrImExSQyVZLopSjkyM8sfRmClKDJ9bgoli9lZEplhhYmw8NRuCt/ryOulftrx2y/Pu2o1Os8yjCmdwDk3w4Ao6cAtd6AGBKTzDK7w5wnlx3p2PZWvFKWdO4Q+czx+kpI5f</latexit>
<latexit
X

k=1
1
=z

X
1

k= 1
X

x[n]
k
n= 1
1
X

ak y[n
a1 x1 [n] + a2 x2 [n] z

n= 1

sha1_base64="uJif1Iw8jM2Z+aSOyInoyCWMiFI=">AAAB63icbVBNS8NAEJ3Ur1q/qh69LBahp5JIoR4LXrwIFa0ttKFstpt26WYTdidCCf0JXjwoiFf/kDf/jds2B219MPB4b4aZeUEihUHX/XYKG5tb2zvF3dLe/sHhUfn45NHEqWa8zWIZ625ADZdC8TYKlLybaE6jQPJOMLme+50nro2I1QNOE+5HdKREKBhFK90Hg9tBueLW3AXIOvFyUoEcrUH5qz+MWRpxhUxSY3qem6CfUY2CST4r9VPDE8omdMR7lioaceNni1Nn5MIqQxLG2pZCslB/T2Q0MmYaBbYzojg2q95c/M/rpRhe+ZlQSYpcseWiMJUEYzL/mwyF5gzl1BLKtLC3EjammjK06ZRsCN7qy+ukc1nz6jXPu6tXmtU8jyKcwTlUwYMGNOEGWtAGBiN4hld4c6Tz4rw7H8vWgpPPnMIfOJ8/o3ONxQ==</latexit>
<latexit sha1_base64="l97gjf8dVW4dgwIJstJJWZqqStQ=">AAAB63icbVBNS8NAEJ3Ur1q/qh69LBahp5JIQY8FLx4rWltoQ9lsJ+3SzSbsboQS+hO8eFAQr/4hb/4bt2kO2vpg4PHeDDPzgkRwbVz32yltbG5t75R3K3v7B4dH1eOTRx2nimGHxSJWvYBqFFxix3AjsJcopFEgsBtMbxZ+9wmV5rF8MLME/YiOJQ85o8ZK98HQHVZrbsPNQdaJV5AaFGgPq1+DUczSCKVhgmrd99zE+BlVhjOB88og1ZhQNqVj7FsqaYTaz/JT5+TCKiMSxsqWNCRXf09kNNJ6FgW2M6Jmole9hfif109NeO1nXCapQcmWi8JUEBOTxd9kxBUyI2aWUKa4vZWwCVWUGZtOxYbgrb68TrqXDa/Z8Ly7Zq1VL/IowxmcQx08uIIW3EIbOsBgDM/wCm+OcF6cd+dj2VpyiplT+APn8wd3Yo2o</latexit>
<latexit
sha1_base64="UJnyHhr0+3uwBtqqUCaFOh8uDyc=">AAAB63icbVBNS8NAEJ3Ur1q/qh69LBahp5KUQj0WvHisaG2hDWWznbRLN5uwuxFK6E/w4kFBvPqHvPlv3LY5aOuDgcd7M8zMCxLBtXHdb6ewtb2zu1fcLx0cHh2flE/PHnWcKoYdFotY9QKqUXCJHcONwF6ikEaBwG4wvVn43SdUmsfywcwS9CM6ljzkjBor3QfD+rBccWvuEmSTeDmpQI72sPw1GMUsjVAaJqjWfc9NjJ9RZTgTOC8NUo0JZVM6xr6lkkao/Wx56pxcWWVEwljZkoYs1d8TGY20nkWB7Yyomeh1byH+5/VTE177GZdJalCy1aIwFcTEZPE3GXGFzIiZJZQpbm8lbEIVZcamU7IheOsvb5JuveY1ap5316i0qnkeRbiAS6iCB01owS20oQMMxvAMr/DmCOfFeXc+Vq0FJ585hz9wPn8AemyNqg==</latexit>
<latexit sha1_base64="FKI7O9vIXsIrLvW979It+DuDBI8=">AAAB63icbVA9SwNBEJ2LXzF+RS1tFoOQKtxKQMuAjWVEYwLJEfY2c8mSvb1jd08IIT/BxkJBbP1Ddv4bN8kVmvhg4PHeDDPzwlQKY33/2ytsbG5t7xR3S3v7B4dH5eOTR5NkmmOLJzLRnZAZlEJhyworsZNqZHEosR2Ob+Z++wm1EYl6sJMUg5gNlYgEZ9ZJ92Gf9ssVv+YvQNYJzUkFcjT75a/eIOFZjMpyyYzpUj+1wZRpK7jEWamXGUwZH7Mhdh1VLEYTTBenzsiFUwYkSrQrZclC/T0xZbExkzh0nTGzI7PqzcX/vG5mo+tgKlSaWVR8uSjKJLEJmf9NBkIjt3LiCONauFsJHzHNuHXplFwIdPXlddK+rNF6jdK7eqVRzfMowhmcQxUoXEEDbqEJLeAwhGd4hTdPei/eu/exbC14+cwp/IH3+QN4542p</latexit>
<latexit
x[n

b0
1

b2
b1

bM
n

h[k]x[n
l= 1

k] +
x2 [n]z

k]
k]z

M
x[l]z

r=0
X
n

sha1_base64="4m043NTs8MYRjIwmVQbJavRyB0M=">AAAB83icbVBNS8NAEJ3Ur1q/qh69BIvgqSSloMeCF48VrC2koWy2m3bpZjfsTpQS+jO8eFAQr/4Zb/4bt20O2vpg4PHeDDPzolRwg5737ZQ2Nre2d8q7lb39g8Oj6vHJg1GZpqxDlVC6FxHDBJesgxwF66WakSQSrBtNbuZ+95Fpw5W8x2nKwoSMJI85JWiloK/5aIxEa/U0qNa8ureAu078gtSgQHtQ/eoPFc0SJpEKYkzgeymGOdHIqWCzSj8zLCV0QkYssFSShJkwX5w8cy+sMnRjpW1JdBfq74mcJMZMk8h2JgTHZtWbi/95QYbxdZhzmWbIJF0uijPhonLn/7tDrhlFMbWEUM3trS4dE00o2pQqNgR/9eV10m3U/Wbd9++atVajyKMMZ3AOl+DDFbTgFtrQAQoKnuEV3hx0Xpx352PZWnKKmVP4A+fzB1hCkbY=</latexit>
<latexit
Z-Transform Pairs and Properties

br x[n
sha1_base64="qdYSLgXjoB6B6jALCluBqv2tcx0=">AAAB63icbVBNS8NAEJ3Ur1q/qh69LBahp5JIoR4LXjxJRWsLbSib7aRdutmE3Y1QQn+CFw8K4tU/5M1/47bNQVsfDDzem2FmXpAIro3rfjuFjc2t7Z3ibmlv/+DwqHx88qjjVDFss1jEqhtQjYJLbBtuBHYThTQKBHaCyfXc7zyh0jyWD2aaoB/RkeQhZ9RY6Z4ObgfliltzFyDrxMtJBXK0BuWv/jBmaYTSMEG17nluYvyMKsOZwFmpn2pMKJvQEfYslTRC7WeLU2fkwipDEsbKljRkof6eyGik9TQKbGdEzVivenPxP6+XmvDKz7hMUoOSLReFqSAmJvO/yZArZEZMLaFMcXsrYWOqKDM2nZINwVt9eZ10LmteveZ5d/VKs5rnUYQzOIcqeNCAJtxAC9rAYATP8ApvjnBenHfnY9lacPKZU/gD5/MHo3GNxQ==</latexit>
<latexit
sha1_base64="SM4VcM2KsiFI4vens7Qc3ltNpX4=">AAAB63icbVBNS8NAEJ3Ur1q/qh69LBahp5KUQj0WvHisaG2hDWWznbRLN5uwuxFK6E/w4kFBvPqHvPlv3LY5aOuDgcd7M8zMCxLBtXHdb6ewtb2zu1fcLx0cHh2flE/PHnWcKoYdFotY9QKqUXCJHcONwF6ikEaBwG4wvVn43SdUmsfywcwS9CM6ljzkjBor3dNhfViuuDV3CbJJvJxUIEd7WP4ajGKWRigNE1Trvucmxs+oMpwJnJcGqcaEsikdY99SSSPUfrY8dU6urDIiYaxsSUOW6u+JjEZaz6LAdkbUTPS6txD/8/qpCa/9jMskNSjZalGYCmJisvibjLhCZsTMEsoUt7cSNqGKMmPTKdkQvPWXN0m3XvMaNc+7a1Ra1TyPIlzAJVTBgya04Bba0AEGY3iGV3hzhPPivDsfq9aCk8+cwx84nz945Y2p</latexit>
<latexit sha1_base64="xiCujF17JFtILLGqNAv+0PRMQx0=">AAAB63icbVA9SwNBEJ2LXzF+RS1tFoOQKtxKQMuAjWVEYwLJEeY2e8mSvb1jd08IIT/BxkJBbP1Ddv4bN8kVmvhg4PHeDDPzwlQKY33/2ytsbG5t7xR3S3v7B4dH5eOTR5NkmvEWS2SiOyEaLoXiLSus5J1Uc4xDydvh+Gbut5+4NiJRD3aS8iDGoRKRYGiddI992i9X/Jq/AFknNCcVyNHsl796g4RlMVeWSTSmS/3UBlPUVjDJZ6VeZniKbIxD3nVUYcxNMF2cOiMXThmQKNGulCUL9ffEFGNjJnHoOmO0I7PqzcX/vG5mo+tgKlSaWa7YclGUSWITMv+bDITmzMqJI8i0cLcSNkKNzLp0Si4EuvryOmlf1mi9RuldvdKo5nkU4QzOoQoUrqABt9CEFjAYwjO8wpsnvRfv3ftYtha8fOYU/sD7/AF3YI2o</latexit>
<latexit
System Input/Output Description sha1_base64="LX1C0jAtkJTIfmjtUpQB2aqfnsE=">AAAB/XicbZDNSsNAFIVv6l+Nf9Eu3QwWQRBLUgq6EQpuXFawttCWMplO2iGTSZiZCCHUV3HjQkHc+h7ufBunbRbaemDg49x7uXeOn3CmtOt+W6W19Y3NrfK2vbO7t3/gHB49qDiVhLZJzGPZ9bGinAna1kxz2k0kxZHPaccPb2b1ziOVisXiXmcJHUR4LFjACNbGGjoVjq6RuAj7fVsY4ugchUOn6tbcudAqeAVUoVBr6Hz1RzFJIyo04VipnucmepBjqRnhdGr3U0UTTEI8pj2DAkdUDfL58VN0apwRCmJpntBo7v6eyHGkVBb5pjPCeqKWazPzv1ov1cHVIGciSTUVZLEoSDnSMZolgUZMUqJ5ZgATycytiEywxESbvGwTgrf85VXo1Gteo+Z5d41qs17kUYZjOIEz8OASmnALLWgDgQye4RXerCfrxXq3PhatJauYqcAfWZ8/ZjyScA==</latexit>
<latexit
a2
a1

aN

r]
let l = n

sha1_base64="ojD/jwiisOdXzsY8gBlGe1XPiok=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBahp5JIoR4LXjxWsLaQhrLZbtqlu5uwuxFC6F/w4kFBvPqDvPlv3LQ5aOuDgcd7M8zMCxPOtHHdb6eytb2zu1fdrx0cHh2f1E/PHnWcKkL7JOaxGoZYU84k7RtmOB0mimIRcjoI57eFP3iiSrNYPpgsoYHAU8kiRrAppMyXwbjecFvuEmiTeCVpQIneuP41msQkFVQawrHWvucmJsixMoxwuqiNUk0TTOZ4Sn1LJRZUB/ny1gW6ssoERbGyJQ1aqr8nciy0zkRoOwU2M73uFeJ/np+a6CbImUxSQyVZLYpSjkyMisfRhClKDM8swUQxeysiM6wwMTaemg3BW395kwyuW1675Xn37Ua3WeZRhQu4hCZ40IEu3EEP+kBgBs/wCm+OcF6cd+dj1Vpxyplz+APn8wemLI5g</latexit>
<latexit
y[n]
! n=l+k
k

3
2

convolution summation
= a1 X1 (z) + a2 X2 (z)

2
10/15/24
 10/15/24

Multiplication
• D: Multiplication
<latexit sha1_base64="WQzQhgFeOTAAgIAHGKrYP+DXOUc=">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</latexit>

y[n] = x1 [n]x2 [n]

X1
n
Y (z) = x1 [n]x2 [n]z
n=1

• We restrict our interest to the unit circle, i.e., z = ej! and assume a Fourier transform
<latexit sha1_base64="AOTjYAUsiRPYYio4jg+BT3q622I=">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</latexit>

1
X
<latexit sha1_base64="GOZwFpyZPns/oAk0DKkKVHvNlDM=">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</latexit>

Y (ej! ) = x1 [n]x2 [n]e j!n

n= 1
X1 ⇢ Z ⇡
1 0 0
= x1 [n] X2 ej! ej! n d! 0 e j!n

n= 1
2⇡ ⇡
Z ⇡ 1
X
1 0
j(! ! 0 )n
= X2 ej! x1 [n]e d! 0
2⇡ ⇡ n= 1
Z ⇡
1 0
!0 )
= X2 ej! X1 ej(! d! 0
2⇡ ⇡

Multiplication
• Multiplication in time domain is convolution is frequency domain, i.e.,
Z ⇡
1
<latexit sha1_base64="UmmUtKyg+QqqD1BEZrO8aHdEcVk=">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</latexit>

0 0
j!
Y (e ) = X2 (ej! )X1 (ej(! ! ) )d! 0
2⇡ ⇡

• Note:
1. This is a periodic or circular convolution (not linear)

Y (ej! ) = X1 (ej! ) X2 (ej! )

<latexit sha1_base64="VlIVHTn2mhCBH9Zas/fsjeP97cY=">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</latexit>

2. Think of circular convolution as a divided cylinder with Fourier transforms pointed on them

3
 10/15/24

Multiplication
• E. Multiplication by an
<latexit sha1_base64="a+yDcCrz84vxDZCh51OOylzHa+E=">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</latexit>

1
X
<latexit sha1_base64="OTxWlcvIHehSS3gI50dhUlI4u/U=">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</latexit>

n
<latexit sha1_base64="OGHI7zxww9d26MeLSA/DAO6Lgas=">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</latexit>

x1 [n] = a x[n] X1 (z) = an x[n] z n

1 n= 1
X1 (z) = X(a z)
X1
1 n
= x[n] a z
let z 0 = a 1
<latexit sha1_base64="dn06GAiqLjuPiJ8EmhU4iIKukzk=">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</latexit>

z
n= 1
X1
= x[n]z 0 n

n= 1

= X(z 0 )
⇣z ⌘
=X
a

Examples
• Let a be real and positive with a  1
<latexit sha1_base64="0aEd6n8+wbSXSIYPnnsevAL9bQQ=">AAAB8HicbVA9SwNBEJ2LXzF+RS1tFoNgFW4loGXAxjKCMYHkCHubuWTJ3t5ld08IIX/CxkJBbP05dv4bN8kVmvhg4PHeDDPzwlQKY33/2ytsbG5t7xR3S3v7B4dH5eOTR5NkmmOTJzLR7ZAZlEJh0worsZ1qZHEosRWObud+6wm1EYl6sJMUg5gNlIgEZ9ZJbUa6EseE9soVv+ovQNYJzUkFcjR65a9uP+FZjMpyyYzpUD+1wZRpK7jEWambGUwZH7EBdhxVLEYTTBf3zsiFU/okSrQrZclC/T0xZbExkzh0nTGzQ7PqzcX/vE5mo5tgKlSaWVR8uSjKJLEJmT9P+kIjt3LiCONauFsJHzLNuHURlVwIdPXlddK6qtJaldL7WqVO8zyKcAbncAkUrqEOd9CAJnCQ8Ayv8OaNvRfv3ftYtha8fOYU/sD7/AGA0Y9i</latexit>

<latexit sha1_base64="LujNQVR6S39KUVtf/B9a3ZVR+as=">AAAB6XicbVA9SwNBEJ2LXzF+RS1tFoNgFW4loGXAxjIBYwLJEfY2c8mavb1jd08IIb/AxkJBbP1Hdv4bN8kVmvhg4PHeDDPzwlQKY33/2ytsbG5t7xR3S3v7B4dH5eOTB5NkmmOLJzLRnZAZlEJhyworsZNqZHEosR2Ob+d++wm1EYm6t5MUg5gNlYgEZ9ZJTdYvV/yqvwBZJzQnFcjR6Je/eoOEZzEqyyUzpkv91AZTpq3gEmelXmYwZXzMhth1VLEYTTBdHDojF04ZkCjRrpQlC/X3xJTFxkzi0HXGzI7MqjcX//O6mY1ugqlQaWZR8eWiKJPEJmT+NRkIjdzKiSOMa+FuJXzENOPWZVNyIdDVl9dJ+6pKa1VKm7VKneZ5FOEMzuESKFxDHe6gAS3ggPAMr/DmPXov3rv3sWwtePnMKfyB9/kDVEmNDQ==</latexit>

• This has the effect of drawing the roots inward on radial paths

4
 10/15/24

Examples
• Let a be complex and on the unit circle, i.e., a = ej!c
<latexit sha1_base64="XBwz06ZZP2wJjY7m+frXgjpPcBk=">AAAB/HicbVBNS8NAEN3Ur1q/Yj16WSyCp5IVQS9CwYvHCtYW2ho220m7djcJuxuxhPwVLx4UxKs/xJv/xm2bg7Y+GHi8N8PMvCARXBvP+3ZKK6tr6xvlzcrW9s7unrtfvdNxqhi0WCxi1QmoBsEjaBluBHQSBVQGAtrB+Grqtx9BaR5Ht2aSQF/SYcRDzqixku9WKb7EcJ894F4sYUh9lvtuzat7M+BlQgpSQwWavvvVG8QslRAZJqjWXeIlpp9RZTgTkFd6qYaEsjEdQtfSiErQ/Wx2e46PrTLAYaxsRQbP1N8TGZVaT2RgOyU1I73oTcX/vG5qwot+xqMkNRCx+aIwFdjEeBoEHnAFzIiJJZQpbm/FbEQVZcbGVbEhkMWXl0n7tE7O6oTcnNUapMijjA7RETpBBJ2jBrpGTdRCDD2hZ/SK3pzceXHenY95a8kpZg7QHzifP2yqk9E=</latexit>

<latexit sha1_base64="LujNQVR6S39KUVtf/B9a3ZVR+as=">AAAB6XicbVA9SwNBEJ2LXzF+RS1tFoNgFW4loGXAxjIBYwLJEfY2c8mavb1jd08IIb/AxkJBbP1Hdv4bN8kVmvhg4PHeDDPzwlQKY33/2ytsbG5t7xR3S3v7B4dH5eOTB5NkmmOLJzLRnZAZlEJhyworsZNqZHEosR2Ob+d++wm1EYm6t5MUg5gNlYgEZ9ZJTdYvV/yqvwBZJzQnFcjR6Je/eoOEZzEqyyUzpkv91AZTpq3gEmelXmYwZXzMhth1VLEYTTBdHDojF04ZkCjRrpQlC/X3xJTFxkzi0HXGzI7MqjcX//O6mY1ugqlQaWZR8eWiKJPEJmT+NRkIjdzKiSOMa+FuJXzENOPWZVNyIdDVl9dJ+6pKa1VKm7VKneZ5FOEMzuESKFxDHe6gAS3ggPAMr/DmPXov3rv3sWwtePnMKfyB9/kDVEmNDQ==</latexit>

• This has the effect of rotating the original z-transform

x1 [n] = an x[n] = ej!n x[n]

⇡
<latexit sha1_base64="u/jl9G5ui1KUiZDsFruPc1mjpEo=">AAACEHicbZDLSgMxFIYzXmu9VV26CRbFVZlIQTdCwY3LCtYWptMhk562sZnMkGSkZegbuPFV3LhQELcu3fk2ppeFtv4Q+PjPOZycP0wE18Z1v52l5ZXVtfXcRn5za3tnt7C3f6fjVDGosVjEqhFSDYJLqBluBDQSBTQKBdTD/tW4Xn8ApXksb80wAT+iXck7nFFjraBwMgiIJ318iWlL4sEUoZXdN+MIuhTL0cQMCkW35E6EF4HMoIhmqgaFr2Y7ZmkE0jBBtfaImxg/o8pwJmCUb6YaEsr6tAueRUkj0H42uWeEj63Txp1Y2ScNnri/JzIaaT2MQtsZUdPT87Wx+V/NS03nws+4TFIDkk0XdVKBTYzH4eA2V8CMGFqgTHH7V8x6VFFmbIR5GwKZP3kR6mclUi4RclMuVsgsjxw6REfoFBF0jiroGlVRDTH0iJ7RK3pznpwX5935mLYuObOZA/RHzucPG5ybOA==</latexit>

<latexit sha1_base64="NbOS1eupwXFzr2Nz1cB9JYDw5CY=">AAACAXicbVA9SwNBEJ2LXzF+nVpY2CwGwSrchoA2QsDGMoIxgdwZ9jZ7yZq9D3b3hHBc41+xsVAQW/+Fnf/GTXKFJj4YeLw3w8w8PxFcacf5tkorq2vrG+XNytb2zu6evX9wp+JUUtamsYhl1yeKCR6xtuZasG4iGQl9wTr++Grqdx6ZVDyObvUkYV5IhhEPOCXaSH37iKBLxO6zB+QGktDMTXie1fO8b1edmjMDWia4IFUo0OrbX+4gpmnIIk0FUaqHnUR7GZGaU8HyipsqlhA6JkPWMzQiIVNeNnsgR6dGGaAglqYijWbq74mMhEpNQt90hkSP1KI3Ff/zeqkOLryMR0mqWUTni4JUIB2jaRpowCWjWkwMIVRycyuiI2KC0CazigkBL768TDr1Gm7UML5pVJu4yKMMx3ACZ4DhHJpwDS1oA4UcnuEV3qwn68V6tz7mrSWrmDmEP7A+fwCRXJYr</latexit>

a = ej 2

Modulation

x[n] is a low pass audio process

<latexit sha1_base64="ZtCBwEVtA7YMeos/Lnxas+3vXgg=">AAAB7HicbVBNSwMxEJ2tX7V+VT16CRbBU9mIoMeCF48VrC1sl5JNs21okl2SrFiW/gUvHhTEqz/Im//GbLsHbX0w8Hhvhpl5USq4sb7/7VXW1jc2t6rbtZ3dvf2D+uHRg0kyTVmHJiLRvYgYJrhiHcutYL1UMyIjwbrR5Kbwu49MG56oeztNWSjJSPGYU2IL6SlQ4aDe8Jv+HGiV4JI0oER7UP/qDxOaSaYsFcSYAPupDXOiLaeCzWr9zLCU0AkZscBRRSQzYT6/dYbOnDJEcaJdKYvm6u+JnEhjpjJynZLYsVn2CvE/L8hsfB3mXKWZZYouFsWZQDZBxeNoyDWjVkwdIVRzdyuiY6IJtS6emgsBL7+8SroXTXzZxPjustHCZR5VOIFTOAcMV9CCW2hDByiM4Rle4c2T3ov37n0sWiteOXMMf+B9/gCnWY5o</latexit>

x1 [n] = an x[n] where a = ej!c

<latexit sha1_base64="o7cJzoPGlR/5uVEn4uLrYsVZZPU=">AAACB3icbVBNS8NAEN3Ur1q/oh5FWCyCp5JIQS9CwYvHCtYWmhg222m7djcJuxuhhJy8+Fe8eFAQr/4Fb/4bN20P2vpg4PHeDDPzwoQzpR3n2yotLa+srpXXKxubW9s79u7erYpTSaFFYx7LTkgUcBZBSzPNoZNIICLk0A5Hl4XffgCpWBzd6HECviCDiPUZJdpIgX1I8AWGu+wee7GAAQkyTxA9lCKjeZ4HdtWpORPgReLOSBXN0AzsL68X01RApCknSnVdJ9F+RqRmlENe8VIFCaEjMoCuoRERoPxs8kaOj43Sw/1Ymoo0nqi/JzIilBqL0HQWN6p5rxD/87qp7p/7GYuSVENEp4v6Kcc6xkUmuMckUM3HhhAqmbkV0yGRhGqTXMWE4M6/vEjapzW3XnPd63q14c7yKKMDdIROkIvOUANdoSZqIYoe0TN6RW/Wk/VivVsf09aSNZvZR39gff4AukmZFA==</latexit>

<latexit sha1_base64="X5M0gNGbq5HkcYY3pOu+5rZteaE=">AAAB+3icbZDNSgMxFIXv1L9a/1pdugkWwVWZiKAboeDGZQVrC+04ZNJMG5rJDElGW8Y+ihsXCuLWF3Hn25i2s9DWA4GPc+/l3pwgEVwb1/12Ciura+sbxc3S1vbO7l65sn+n41RR1qSxiFU7IJoJLlnTcCNYO1GMRIFgrWB4Na23HpjSPJa3ZpwwLyJ9yUNOibGWX66MfNyRHrpE5F6ikUW/XHVr7kxoGXAOVcjV8Mtf3V5M04hJQwXRuoPdxHgZUYZTwSalbqpZQuiQ9FnHoiQR0142O32Cjq3TQ2Gs7JMGzdzfExmJtB5Hge2MiBnoxdrU/K/WSU144WVcJqlhks4XhalAJkbTHFCPK0aNGFsgVHF7K6IDogg1Nq2SDQEvfnkZWqc1fFbD+OasWsd5HkU4hCM4AQznUIdraEATKDzCM7zCm/PkvDjvzse8teDkMwfwR87nDzHAkxM=</latexit>

amplitude modulation x1 [n] = sin(!c n)x[n]

<latexit sha1_base64="E5bX1GzXpXSBFwIA/KjkZ2+QjIU=">AAACJHicbVBNSwMxEM36bf2qevQSLIJeykYLiiAUvHhUsLbQXZZsOm2DSXZJstKy7H/x4l/x4kFRPHjxt5jWHtT6YODx3gwz8+JUcGN9/8ObmZ2bX1hcWi6trK6tb5Q3t25MkmkGDZaIRLdiakBwBQ3LrYBWqoHKWEAzvj0f+c070IYn6toOUwgl7Sne5YxaJ0Xl00FE2irEZzgwXOF9HCQSejTKA0ltX8ucFQVW+AAHfZNSBnn1SMoCD9xMVK74VX8MPE3IhFTQBJdR+TXoJCyToCwT1Jg28VMb5lRbzgQUpSAz4Hbc0h60HVVUggnz8Y8F3nNKB3cT7UpZPFZ/TuRUGjOUsescXW7+eiPxP6+d2e5JmHOVZhYU+17UzQS2CR4FhjtcA7Ni6AhlmrtbMetTTZl1sZZcCOTvy9OkeVgltSohV7VKnUzyWEI7aBftI4KOUR1doEvUQAzdo0f0jF68B+/Je/Pev1tnvMnMNvoF7/MLxfGjhA==</latexit>

5
 10/15/24

Modulation

x[n] is a low pass audio process

<latexit sha1_base64="ZtCBwEVtA7YMeos/Lnxas+3vXgg=">AAAB7HicbVBNSwMxEJ2tX7V+VT16CRbBU9mIoMeCF48VrC1sl5JNs21okl2SrFiW/gUvHhTEqz/Im//GbLsHbX0w8Hhvhpl5USq4sb7/7VXW1jc2t6rbtZ3dvf2D+uHRg0kyTVmHJiLRvYgYJrhiHcutYL1UMyIjwbrR5Kbwu49MG56oeztNWSjJSPGYU2IL6SlQ4aDe8Jv+HGiV4JI0oER7UP/qDxOaSaYsFcSYAPupDXOiLaeCzWr9zLCU0AkZscBRRSQzYT6/dYbOnDJEcaJdKYvm6u+JnEhjpjJynZLYsVn2CvE/L8hsfB3mXKWZZYouFsWZQDZBxeNoyDWjVkwdIVRzdyuiY6IJtS6emgsBL7+8SroXTXzZxPjustHCZR5VOIFTOAcMV9CCW2hDByiM4Rle4c2T3ov37n0sWiteOXMMf+B9/gCnWY5o</latexit>

x1 [n] = an x[n] where a = ej!c

<latexit sha1_base64="o7cJzoPGlR/5uVEn4uLrYsVZZPU=">AAACB3icbVBNS8NAEN3Ur1q/oh5FWCyCp5JIQS9CwYvHCtYWmhg222m7djcJuxuhhJy8+Fe8eFAQr/4Fb/4bN20P2vpg4PHeDDPzwoQzpR3n2yotLa+srpXXKxubW9s79u7erYpTSaFFYx7LTkgUcBZBSzPNoZNIICLk0A5Hl4XffgCpWBzd6HECviCDiPUZJdpIgX1I8AWGu+wee7GAAQkyTxA9lCKjeZ4HdtWpORPgReLOSBXN0AzsL68X01RApCknSnVdJ9F+RqRmlENe8VIFCaEjMoCuoRERoPxs8kaOj43Sw/1Ymoo0nqi/JzIilBqL0HQWN6p5rxD/87qp7p/7GYuSVENEp4v6Kcc6xkUmuMckUM3HhhAqmbkV0yGRhGqTXMWE4M6/vEjapzW3XnPd63q14c7yKKMDdIROkIvOUANdoSZqIYoe0TN6RW/Wk/VivVsf09aSNZvZR39gff4AukmZFA==</latexit>

<latexit sha1_base64="X5M0gNGbq5HkcYY3pOu+5rZteaE=">AAAB+3icbZDNSgMxFIXv1L9a/1pdugkWwVWZiKAboeDGZQVrC+04ZNJMG5rJDElGW8Y+ihsXCuLWF3Hn25i2s9DWA4GPc+/l3pwgEVwb1/12Ciura+sbxc3S1vbO7l65sn+n41RR1qSxiFU7IJoJLlnTcCNYO1GMRIFgrWB4Na23HpjSPJa3ZpwwLyJ9yUNOibGWX66MfNyRHrpE5F6ikUW/XHVr7kxoGXAOVcjV8Mtf3V5M04hJQwXRuoPdxHgZUYZTwSalbqpZQuiQ9FnHoiQR0142O32Cjq3TQ2Gs7JMGzdzfExmJtB5Hge2MiBnoxdrU/K/WSU144WVcJqlhks4XhalAJkbTHFCPK0aNGFsgVHF7K6IDogg1Nq2SDQEvfnkZWqc1fFbD+OasWsd5HkU4hCM4AQznUIdraEATKDzCM7zCm/PkvDjvzse8teDkMwfwR87nDzHAkxM=</latexit>

amplitude modulation x1 [n] = sin(!c n)x[n]

<latexit sha1_base64="E5bX1GzXpXSBFwIA/KjkZ2+QjIU=">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</latexit>

10

10

Modulation

x[n] is a low pass audio process

<latexit sha1_base64="ZtCBwEVtA7YMeos/Lnxas+3vXgg=">AAAB7HicbVBNSwMxEJ2tX7V+VT16CRbBU9mIoMeCF48VrC1sl5JNs21okl2SrFiW/gUvHhTEqz/Im//GbLsHbX0w8Hhvhpl5USq4sb7/7VXW1jc2t6rbtZ3dvf2D+uHRg0kyTVmHJiLRvYgYJrhiHcutYL1UMyIjwbrR5Kbwu49MG56oeztNWSjJSPGYU2IL6SlQ4aDe8Jv+HGiV4JI0oER7UP/qDxOaSaYsFcSYAPupDXOiLaeCzWr9zLCU0AkZscBRRSQzYT6/dYbOnDJEcaJdKYvm6u+JnEhjpjJynZLYsVn2CvE/L8hsfB3mXKWZZYouFsWZQDZBxeNoyDWjVkwdIVRzdyuiY6IJtS6emgsBL7+8SroXTXzZxPjustHCZR5VOIFTOAcMV9CCW2hDByiM4Rle4c2T3ov37n0sWiteOXMMf+B9/gCnWY5o</latexit>

x1 [n] = an x[n] where a = ej!c

<latexit sha1_base64="o7cJzoPGlR/5uVEn4uLrYsVZZPU=">AAACB3icbVBNS8NAEN3Ur1q/oh5FWCyCp5JIQS9CwYvHCtYWmhg222m7djcJuxuhhJy8+Fe8eFAQr/4Fb/4bN20P2vpg4PHeDDPzwoQzpR3n2yotLa+srpXXKxubW9s79u7erYpTSaFFYx7LTkgUcBZBSzPNoZNIICLk0A5Hl4XffgCpWBzd6HECviCDiPUZJdpIgX1I8AWGu+wee7GAAQkyTxA9lCKjeZ4HdtWpORPgReLOSBXN0AzsL68X01RApCknSnVdJ9F+RqRmlENe8VIFCaEjMoCuoRERoPxs8kaOj43Sw/1Ymoo0nqi/JzIilBqL0HQWN6p5rxD/87qp7p/7GYuSVENEp4v6Kcc6xkUmuMckUM3HhhAqmbkV0yGRhGqTXMWE4M6/vEjapzW3XnPd63q14c7yKKMDdIROkIvOUANdoSZqIYoe0TN6RW/Wk/VivVsf09aSNZvZR39gff4AukmZFA==</latexit>

<latexit sha1_base64="X5M0gNGbq5HkcYY3pOu+5rZteaE=">AAAB+3icbZDNSgMxFIXv1L9a/1pdugkWwVWZiKAboeDGZQVrC+04ZNJMG5rJDElGW8Y+ihsXCuLWF3Hn25i2s9DWA4GPc+/l3pwgEVwb1/12Ciura+sbxc3S1vbO7l65sn+n41RR1qSxiFU7IJoJLlnTcCNYO1GMRIFgrWB4Na23HpjSPJa3ZpwwLyJ9yUNOibGWX66MfNyRHrpE5F6ikUW/XHVr7kxoGXAOVcjV8Mtf3V5M04hJQwXRuoPdxHgZUYZTwSalbqpZQuiQ9FnHoiQR0142O32Cjq3TQ2Gs7JMGzdzfExmJtB5Hge2MiBnoxdrU/K/WSU144WVcJqlhks4XhalAJkbTHFCPK0aNGFsgVHF7K6IDogg1Nq2SDQEvfnkZWqc1fFbD+OasWsd5HkU4hCM4AQznUIdraEATKDzCM7zCm/PkvDjvzse8teDkMwfwR87nDzHAkxM=</latexit>

amplitude modulation x1 [n] = cos(!c n)x[n]

<latexit sha1_base64="UJ47KRyGzR3HX13+vSrIe+ord90=">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</latexit>

11

11

6
 10/15/24

Complex Basebanding

frequency content of signal of interest

!c = 2⇡fc
<latexit sha1_base64="Gdmr4xjLgJBQAiW36vcRmmRqMb0=">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</latexit>

!c j! n
<latexit sha1_base64="TlxdDIIDBPJwtu1Ek54i94ula0w=">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</latexit>

multiplication by e
<latexit sha1_base64="mHs8ZmIEw4ChqP/h+v3KyqMjBa4=">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</latexit>

results in a clockwise rotation

of z-transform

12

12

Complex Basebanding

frequency content of signal of interest

!c = 2⇡fc
<latexit sha1_base64="Gdmr4xjLgJBQAiW36vcRmmRqMb0=">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</latexit>

!c j! n
<latexit sha1_base64="TlxdDIIDBPJwtu1Ek54i94ula0w=">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</latexit>

multiplication by e
<latexit sha1_base64="mHs8ZmIEw4ChqP/h+v3KyqMjBa4=">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</latexit>

results in a clockwise rotation

of z-transform

13

13

7
 10/15/24

Z-Transform of Finite Length Sequences

N =4
<latexit sha1_base64="P1y+ylIPRZ1nPqG+wJ44hkZceS4=">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</latexit>

a=1

• Assume sequence is of length N

<latexit sha1_base64="llTmgWDynac/KlsaU/YYC5tpVaM=">AAACEnicbVDLSsNAFJ34rPFVdekmWAQXpUy0oMuiC11JK9YW2lIm05t26GQSZiZCCfkCl+rHuBS3/oDf4sZpmoW2Hhg4nHMv98zxIs6UxvjLWlpeWV1bL2zYm1vbO7vFvf0HFcaSQpOGPJRtjyjgTEBTM82hHUkggceh5Y2vpn7rEaRiobjXkwh6ARkK5jNKtJEat/1iCVdwBmeRuDkpoRz1fvG7OwhpHIDQlBOlOi6OdC8hUjPKIbW7sYKI0DEZQsdQQQJQvSQLmjrHRhk4fijNE9rJ1N8bCQmUmgSemQyIHql5byr+53Vi7V/0EiaiWIOgs0N+zB0dOtNfOwMmgWo+MYRQyUxWh46IJFSbbuzuAHzTXxYnUZGphvChBBBpcnd9mSa47J7hMk5t05Y7380iaZ1W3GrFdRvVUg3nxRXQITpCJ8hF56iGblAdNRFFgJ7QC3q1nq036936mI0uWfnOAfoD6/MHLFac7Q==</latexit>

N =4
<latexit sha1_base64="PYZYlP1u/o1iEGZL4CaZDhreNTo=">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</latexit>

N
X1
<latexit sha1_base64="tPAgHBYqPvlQrDHupnVgiIRBt0A=">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</latexit>

a = 0.9
n
x[n] = an for n = 0, . . . , N
<latexit sha1_base64="T5eeEy6qzhf4XydQn7oxQ9TQ+JI=">AAACG3icbVDLSgMxFM34rPVVdekmWAQXpWS0oBuh6EKXVawttmPJpHdqaCYzJBmxDPMXLtWPcSluXfgtbkwfC60eCBzOuZd7cvxYcG0I+XRmZufmFxZzS/nlldW19cLG5rWOEsWgziIRqaZPNQguoW64EdCMFdDQF9Dw+6dDv3EPSvNIXplBDF5Ie5IHnFFjpZuHlvTwMaa3slMokjIZAf8l7oQU0QS1TuGr3Y1YEoI0TFCtWy6JjZdSZTgTkOXbiYaYsj7tQctSSUPQXjpKnOFdq3RxECn7pMEj9edGSkOtB6FvJ0Nq7vS0NxT/81qJCY68lMs4MSDZ+FCQCGwiPPw+7nIFzIiBJZQpbrNidkcVZcaWlG93IbBFjuKkOrYdUdFTADJLL89OspSU3ANSIlnetuVOd/OXNPbLbqXsuheVYpVMisuhbbSD9pCLDlEVnaMaqiOGJHpEz+jFeXJenTfnfTw640x2ttAvOB/fibagQQ==</latexit>

X(z) = x[n]z 1
<latexit sha1_base64="fTgT6r9dETOwjHInhoYkbsUbSec=">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</latexit>

n=0
N
X1
<latexit sha1_base64="Xjdkj2Wdk+x0YrvwXuYZdc92qkg=">AAACrnicbVFda9swFJW9j3bZV7o97kUsjCXQZFJb2F4KZXvYnkI3lqUQO0ZWZEdUlo0kDxKhX9Jf1t+ylym2KV27C4Kjc87VFeemleDaIHQdhA8ePnq8t/+k9/TZ8xcv+wevfumyVpTNaClKdZESzQSXbGa4EeyiUowUqWDz9PLLTp//ZkrzUv40m4rFBcklzzglxlNJ/yrSdbG00zF2iZWnyMGIiGpNlhKewihThFo87qipu8Huljok26X1/aPWcHNtPCnP82Hr3Dq7HRPXUKNO6R7ZLqdwDJsJHrpWHCX9AZqgpuB9gDswAF2dJ/0/0aqkdcGkoYJovcCoMrElynAqmOtFtWYVoZckZwsPJSmYjm0TooPvPLOCWan8kQY27O0OSwqtN0XqnQUxa31X25H/0xa1yT7FlsuqNkzSdlBWC2hKuNsIXHHFqBEbDwhV3P8V0jXxsRi/t160YpnfbfMdqyu/NiJyxZh09sfXz86iQ3yMDpHr+bTw3Wzug/nRBJ9MMP5+MjhDXXD74A14C4YAg4/gDHwD52AGaBAE7wMU4PBDOAvjMGmtYdD1vAb/VLj+CxqCz5k=</latexit>

✓ ◆✓ ◆
1 ↵N 1 (az 1 )N n z zN aN
↵ = = =
n=0
1 ↵ 1 (az 1 ) z a zN

14

14

Z-Transform of Finite Length Sequences

✓ ◆✓ ◆
zN aN
<latexit sha1_base64="KwxXlnlyi7NpELkWrAf55wWr16U=">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</latexit>

z
x[n] = an for n = 0, . . . , N
<latexit sha1_base64="T5eeEy6qzhf4XydQn7oxQ9TQ+JI=">AAACG3icbVDLSgMxFM34rPVVdekmWAQXpWS0oBuh6EKXVawttmPJpHdqaCYzJBmxDPMXLtWPcSluXfgtbkwfC60eCBzOuZd7cvxYcG0I+XRmZufmFxZzS/nlldW19cLG5rWOEsWgziIRqaZPNQguoW64EdCMFdDQF9Dw+6dDv3EPSvNIXplBDF5Ie5IHnFFjpZuHlvTwMaa3slMokjIZAf8l7oQU0QS1TuGr3Y1YEoI0TFCtWy6JjZdSZTgTkOXbiYaYsj7tQctSSUPQXjpKnOFdq3RxECn7pMEj9edGSkOtB6FvJ0Nq7vS0NxT/81qJCY68lMs4MSDZ+FCQCGwiPPw+7nIFzIiBJZQpbrNidkcVZcaWlG93IbBFjuKkOrYdUdFTADJLL89OspSU3ANSIlnetuVOd/OXNPbLbqXsuheVYpVMisuhbbSD9pCLDlEVnaMaqiOGJHpEz+jFeXJenTfnfTw640x2ttAvOB/fibagQQ==</latexit>

X(z) = 1
<latexit sha1_base64="fTgT6r9dETOwjHInhoYkbsUbSec=">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</latexit>

z a zN

zeros are equally

spaced around
circle of radius a
<latexit sha1_base64="v/GfaNkBcYpcG5ZQzKus0raSGOA=">AAACEnicbVDLSsNAFJ3UV42vqks3wSK4KGWiBV0WXeiyFWsLbSmT6U07dDIJMxOhhHyBS/VjXIpbf8BvceM0zUJbDwwczrmXe+Z4EWdKY/xlFVZW19Y3ipv21vbO7l5p/+BBhbGk0KIhD2XHIwo4E9DSTHPoRBJI4HFoe5Prmd9+BKlYKO71NIJ+QEaC+YwSbaQmGZTKuIozOMvEzUkZ5WgMSt+9YUjjAISmnCjVdXGk+wmRmlEOqd2LFUSETsgIuoYKEoDqJ1nQ1DkxytDxQ2me0E6m/t5ISKDUNPDMZED0WC16M/E/rxtr/7KfMBHFGgSdH/Jj7ujQmf3aGTIJVPOpIYRKZrI6dEwkodp0Y/eG4Jv+sjiJikw1hI8kgEiTu5urNMEV9xxXcGqbttzFbpZJ+6zq1qqu26yV6zgvroiO0DE6RS66QHV0ixqohSgC9IRe0Kv1bL1Z79bHfLRg5TuH6A+szx9MQJ0A</latexit>

15

15

8
 10/15/24

Z-Transform of Finite Length Sequences

16

16

9
Z-Transform Definition

∞
X
X (z) = x[n]z −n
n=−∞

Will use the notation X (z) = Z(x[n]).

Definition not complete without specifying the ROC
∞
X ∞
X
−n
ROC : {z : |x[n]z |= |x[n]||z|−n < ∞}
n=−∞ n=−∞

Why Z -transforms?
▶ More general: Applicable to a larger class of signals
▶ Easier to manipulate: Complex variable theory can be useful
(Laurent series and associated results).
Since |e jω | = 1, DTFT is the z-transform evaluated on the unit circle.
General Sequences

1. Finite Duration Sequences. ROC: All z except possibly z = 0 or

z =∞
2. Right Sided Sequences. ROC: {z : |z| > rR } except possibly z = ∞
3. Left Sided Sequences. ROC: {z : |z| < rL } except possibly z = 0
4. Two Sided Sequences. ROC: {z : rR < |z| < rL } is an annular ring
Z-Transforms and Difference Equations
Y (z)
y [n] = h[n] ∗ x[n] or Y (z) = H(z)X (z) leading to H(z) = X (z) , the
transfer function.
Difference equation description of a LTI system
N M N M
X X X ak X bm
ak y [n−k] = bm x[n−m] or y [n] = − y [n−k]+ x[n−m]
m=0
a0 a
m=0 0
k=0 k=1
Taking Z-transform of both sides
N
! M
!
X X
Z ak y [n − k] = Z bm x[n − m]
k=0 m=0
N
X M
X
ak Z(y [n − k]) = bm Z(x[n − m])
k=0 m=0
N
X M
X
ak z −k Y (z) = bm z −m X (z)
k=0 m=0
N
! M
!
X X
−k −m
ak z Y (z) = bm z X (z)
k=0 m=0
 N
X M
X
ak y [n − k] = bm x[n − m]
k=0 m=0
N
! M
!
X X
Z ak y [n − k] = Z bm x[n − m]
k=0 m=0
N
! M
!
X X
ak z −k Y (z) = bm z −m X (z)
k=0 m=0
A(z)Y (z) = B(z)X (z)
PN −k
PM
where A(z) = k=0 ak z and B(z) = m=0 bm z −m .
PN
Y (z) B(z) ak z −k
H(z) = = = PMk=0
X (z) A(z) m=0 bm z
−m

There is a simple transformation from a difference equation to a transfer

function and vice versa. Should be able to go back and forth between the
two effortlessly.
Examples

Difference Equation to Transfer Function

2 + 3z −3
y [n]−y [n −1]+.8y [n −2] = 2x[n]+3x[n −3] ↔ H(z) =
1 − z −1 + .8z −2
Transfer Function to Difference Equation

1 + 2z −2 + z −4
H(z) =
1 − 1.2z −1 + .75z −2
↔ y [n] − 1.2y [n − 1] + .75y [n − 2] = x[n] + 2x[n − 2] + x[n − 4]
Stability and Causality

PM
Y (z) B(z) bm z −m b0 Π M
m=1 (1 − zm z
−1
)
H(z) = = = Pm=0
N
= N
X (z) A(z) −k a −1
P
k=0 ak z 0 k=1 (1 − pk z )

Poles determine ROC. Zeros do not impact ROC

Causality: ROC: {z : |z| > maxk |pk |}

Stability: ROC: Region between poles, but not including any poles, that
contains the unit circle

Causal and Stable: ROC: {z : |z| > maxk |pk |} and include unit circle.

Need maxk |pk | to be less than 1. All poles must be inside the unit circle
Example

Causal: ROC is {z : |z| > 2}

1
Stable system: ROC is {z : 2 < |z| < 2}

Causal and Stable: Not possible because all the poles are not inside the
unit circle.
Inverse Z-Transform

Given X (z) and associated ROC, what is x[n]?

1. By inspection
Example: H(z) = 1 + 2z −3 ↔ h[n] = δ[n] + 2δ[n − 3]
PM PM
Example: H(z) = m=0 bm z −m ↔ h[n] = m=0 bm δ[n − m]

2. By Partial Fraction Expansion and Table Lookup

PM M−N
bm z −m X P(z)
X (z) = PN
m=0
= br z −r +
k=0 ak z
−k
r =0
A(z)
M−N N
X X Ak
= br z −r +
r =0
1 − pk z −1
k=1

Here the assumptions are M ≥ N and simple roots. Note P(z) is of order
(N − 1).
Magnitude, Phase, Group-Delay of Difference Equations
Based LTI systems
PM −jωm −jω
jωB(e jω ) m=0 bm e b0 Π M
m=1 (1 − zm e )
H(e ) = jω
= N
= N −jω
A(e ) −jωk a0 Πk=1 (1 − pk e )
P
k=0 ak e
Magnitude Response
M
b0 X
20 log10 |H(e jω )| = 20 log10 | |+ 20 log10 |1 − zm e −jω |
a0 m=1
N
X
− 20 log10 |1 − pk e −jω |
k=1

Phase
M N
b0 X X
ϕ(ω) = ∠H(e jω ) = ∠ + ∠(1 − zm e −jω ) − ∠(1 − pk e −jω )
a0 m=1
k=1
Group-Delay
M N
dϕ(ω) X d∠(1 − zm e −jω ) X d∠(1 − pk e −jω )
τ (ω) = − =− +
dω m=1
dω dω
k=1
First Order Systems

Let zm = re jθ . Then 1 − zm e −jω = 1 − re −j(ω−θ) . Noting that |a|2 = aa∗ ,

we have the magnitude response

|1 − zk e −jω |2 = (1 − re −j(ω−θ) )(1 − re j(ω−θ) ) = 1 + r 2 − 2r cos(ω − θ)

20 log10 |1 − zk e −jω | = 10 log10 (1 + r 2 − 2r cos(ω − θ))

Phase: 1 − re −j(ω−θ) = 1 − r cos(ω − θ) + jr sin(ω − θ).

r sin(ω − θ)
ARG(1 − re −j(ω−θ) ) = arctan
1 − r cos(ω − θ)
First Order System

e jω − re jθ v3
H(e jω ) = 1 − re jθ e −jω = jω
=
e v1
|v3 |
|H(e jω )| = and ∠H(e jω ) = ∠v3 − ∠v1 = ϕ3 − ω
|v1 |
Second Order System

1 e j2ω v32
H(e jω ) = = =
(1 − re jθ e −jω )(1 − re −jθ e −jω ) (e jω − re jθ )(e jω − re −jθ ) v1 v2

|v3 |2
|H(e jω )| = and ∠H(e jω ) = 2∠v3 − ∠v1 − ∠v2
|v1 ||v2 |
Implementation of Filters
Start with the difference equation
N M
y[ n ] = ∑a
k =1
k y[n − k ] + ∑ bk x[n − k ]
k =0

Corresponding transfer function

∑k =0 k −k
M
B( z ) b z
H ( z) = =
A( z ) 1 − ∑ ak z −k
N
k =1
Hardware Requirements

 Adders
 Multipliers
 Storage
Direct Form I

B(z)   1/A(z)
 Direct Form II

1/A(z) B(z)
Figure 6.8 Example of nodes and branches in a signal flow graph.

Signal Flow Graph: 1) a network of directed branches that

connect at nodes. 2) Each node is associated with a variable.
3) Each branch has an input value and an output values.
Figure 6.9 Example of a signal flow graph showing source and sink nodes.
Figure 6.10 (a) Block diagram representation of a 1st-order digital filter. (b) Structure of the signal flow graph
corresponding to the block diagram in (a).
Figure 6.11 Signal flow graph of Figure 6.10(b) with the delay branch indicated by z−1.
Direct Form I (Signal Flow Graph)
Direct Form II (Signal Flow Graph)
 −1 −2
Example: 1 + 2z + z
H ( z) = −1 −2
1 − .75z + .125z

Direct Form I

Direct Form II
Transposed Form
1. Reverse the direction of all branches
2. Interchange input and Output
Finite Precision Arithmetic
• Coefficient Quantization
• Representing the coefficients of the filters using a finite number of bits and
the related degradation in filter transfer function

• Roundoff Noise
• Relates to the errors resulting from the arithmetic operations carried out in
finite precision
• Nature of the errors depend on fixed point arithmetic versus floating point
arithmetic
Coefficient Quantization
∑k =0 bk z −k
M
 ˆ
H ( z) =
1 − ∑k =1 aˆ k z −k
N

 Where ak = ak + ∆ak and k = bk + ∆bk

ˆ ˆ
b
 For coefficient sensitivity, we need to
understand how errors in coefficients
affect the zeros and poles
N
A( z ) = 1 − ∑ ak z −k
= ∏l =1 (1 − zl z −1 )
N

k =1
N
∂zi
∆zi = ∑ ∆ak , i = 1,2,.. N
k =1 ∂ak
 Coefficient Quantization Cont’d

∂zi zi
=
∂ak ∏
N
j =1, j ≠i
( zi − z j )

Conclusion: Closely spaced roots is a problem. Small perturbation

in coefficients can cause large perturbation of roots.
Table 6.1 UNQUANTIZED DIRECT-FORM COEFFICIENTS FOR A 12TH-ORDER ELLIPTIC FILTER
Figure 6.48 IIR coefficient quantization example. (a) Poles and zeros of H(z) for unquantized
coefficients. (b) Poles and zeros for 16-bit quantization of the direct form coefficients. (Bandpass Filter)
 Pz
Pz X
X
p xpI

pi p
p2 X Xl
X
pz

Filter 1 Filter 2
 Pz
X
xp

xx
pg

Filter 2

Pz
X
Xp

P
X
pz

Implement Pole1 Implement Pole 2

Cascade Form
N
b0 k + b1k z −1 + b2 k z −2
H ( z) = ∏ −1 −2
k =1 1 − a1k z − a2 k z
Parallel Form
−1
e Ns
+ e z
H ( z) = ∑ C z + ∑ −k
Np 0k 1k
−1 −2
k =0
1− a −
k
k =0 1k z a 2k z
Figure 6.47 IIR coefficient quantization example. (a) Log magnitude for unquantized elliptic bandpass filter.
(b) Magnitude in passband for unquantized (solid line) and 16-bit quantized cascade form (dashed line).
 X [11]

X [n] y[n] /'CO 0

2r cos 0 -r in 8

y[11)
/'CO 0 �
_,,2

z-planc

---
!J,m z-plane
1.00 o Realizable pole positions
o Realizable pole positions
1.00 -- -
--- r--_
r--. .___
0.75 r--,
0.75 Unit circle
''
0.50

0.50

0.25

0.25

0 0.25 0.50 'l./1,e

(a)

z-planc
0 0.25 0.50 0.7:> 1.00

1.0 (a)

z-plane

1.0.----

0.5

0.5

0 0.5 1. 0 2/l.e
(b)
Figure 6.42 Pole-locations for the
second-order IIR direct-form system of
Figure 6.41. (a) Four-bit quantization of 0 0.5 1.0 'llle
coefficients. (b) Seven-bit quantization.
(b)
Z-Plane Pole Locations and Their Corresponding Time Series

0
Z-Plane Pole Locations and Their Corresponding Time Series

1
 10/17/24

Z-Transform of Sinusoidal Sequence

• We consider the sinusoidal sequence x[n] = (sin !n)u[n] ; note that
<latexit sha1_base64="7pK+kuP3x2FjFQN8FefDRrlmTzE=">AAACE3icbVDLSgMxFM3UV62vUZdugkVoN2VGC7oRCm5cVrC20BlKJr1tQ5PMkGTEMvQb3PgrblwoiFs37vwb08dCWw8EDuecy809UcKZNp737eRWVtfWN/Kbha3tnd09d//gTsepotCgMY9VKyIaOJPQMMxwaCUKiIg4NKPh1cRv3oPSLJa3ZpRAKEhfsh6jxFip45Yf2jLEl7gUaCZxEAvoEyzLOBjohFDIKr4QY5zaUMctehVvCrxM/DkpojnqHfcr6MY0FSAN5UTrtu8lJsyIMoxyGBeCVIPdMSR9aFsqiQAdZtOTxvjEKl3ci5V90uCp+nsiI0LrkYhsUhAz0IveRPzPa6emdxFmTCapAUlni3opxybGk35wlymgho8sIVQx+1dMB0QRamyLBVuCv3jyMmmeVvxqxfdvqsXa2byPPDpCx6iEfHSOauga1VEDUfSIntErenOenBfn3fmYRXPOfOYQ/YHz+QNfVpxw</latexit>

sin !n = ej!n j!n

<latexit sha1_base64="jPcaW1nGWLOeb7iM0GvhZeI9UlE=">AAACRXicbVDLSgMxFM34tr6qLt0Ei1AXrTMq6EYouHFZwVqhU0smvW1jk8yQZIQyzMe58QPc+QVuXCiIW83UIrb1QuDknHPvTU4QcaaN6z47M7Nz8wuLS8u5ldW19Y385ta1DmNFoUZDHqqbgGjgTELNMMPhJlJARMChHvTPM71+D0qzUF6ZQQRNQbqSdRglxlKtfMPXTGI/FNAlWOIz7AesW4Tb5O6XTLHf0xGhkJTKR0KkpfErtubSH3c2YB8f4MO7Vr7glt1h4WngjUABjarayj/57ZDGAqShnGjd8NzINBOiDKMc0pwfa7Cb+6QLDQslEaCbyTCEFO9Zpo07obJHGjxk/3YkRGg9EIF1CmJ6elLLyP+0Rmw6p82EySg2IOnPok7MsQlxlihuMwXU8IEFhCpm34ppjyhCjc09Z0PwJr88DeqHZe+47HmXx4XK0SiPJbSDdlEReegEVdAFqqIaougBvaA39O48Oq/Oh/P5Y51xRj3baKycr28sjq97</latexit>

e /2j
<latexit sha1_base64="+Pp0QaiVLP33abn3mgK7tHwmLqY=">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</latexit>

1
X
n
X(z) = sin[!n]z
n=0
X1 ✓ jwn jwn
◆
e e n
= z
n=0
2j
1 1
1 X j!n n 1 X j!n n
= e z e z
2j n=0 2j n=0
✓ ◆ ✓ ◆
1 z 1 z
=
2j z ej! 2j z e j!
z sin !
= |z| > 1
<latexit sha1_base64="p0Q8/8cIinMl3AhZ7t2jCUYgVwI=">AAAB7HicbVBNSwMxEJ2tX7V+VT16CRbBU9kVqZ6k4MVjBbcttEvJptk2NJssSVao2/4GLx4U8eoP8ua/MW33oK0PBh7vzTAzL0w408Z1v53C2vrG5lZxu7Szu7d/UD48amqZKkJ9IrlU7RBrypmgvmGG03aiKI5DTlvh6Hbmtx6p0kyKBzNOaBDjgWARI9hYyZ88TW68XrniVt050CrxclKBHI1e+avblySNqTCEY607npuYIMPKMMLptNRNNU0wGeEB7VgqcEx1kM2PnaIzq/RRJJUtYdBc/T2R4VjrcRzazhiboV72ZuJ/Xic10XWQMZGkhgqyWBSlHBmJZp+jPlOUGD62BBPF7K2IDLHCxNh8SjYEb/nlVdK8qHq1au3+slJ38ziKcAKncA4eXEEd7qABPhBg8Ayv8OYI58V5dz4WrQUnnzmGP3A+fwCf4o6G</latexit>

(z e )(z e j! )
j!

Interference Cancellation
x0 (t)
<latexit sha1_base64="0hw/8sX807CChFbLeVs7CgIU1z4=">AAAB7XicbVBNS8NAEJ34WetX1aOXxSJWkJJIQY8FLx4rGFtoQ9lsN+3SzSbsTsRS+hu8eFAQr/4fb/4bt20O2vpg4PHeDDPzwlQKg6777aysrq1vbBa2its7u3v7pYPDB5NkmnGfJTLRrZAaLoXiPgqUvJVqTuNQ8mY4vJn6zUeujUjUPY5SHsS0r0QkGEUr+U9nFTzvlspu1Z2BLBMvJ2XI0eiWvjq9hGUxV8gkNabtuSkGY6pRMMknxU5meErZkPZ521JFY26C8ezYCTm1So9EibalkMzU3xNjGhszikPbGVMcmEVvKv7ntTOMroOxUGmGXLH5oiiTBBMy/Zz0hOYM5cgSyrSwtxI2oJoytPkUbQje4svLpHlZ9WpVz7urlesXeR4FOIYTqIAHV1CHW2iADwwEPMMrvDnKeXHenY9564qTzxzBHzifP3MujjM=</latexit>

<latexit sha1_base64="teIAQSHnq3BmnosM5F+tbG+jXjA=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECnosePFYwdpCGspmu2mX7m7C7kYsoX/BiwcF8eoP8ua/cZPmoK0PBh7vzTAzL0w408Z1v53K2vrG5lZ1u7azu7d/UD88etBxqgjtkpjHqh9iTTmTtGuY4bSfKIpFyGkvnN7kfu+RKs1ieW9mCQ0EHksWMYJNLj35MhjWG27TLYBWiVeSBpToDOtfg1FMUkGlIRxr7XtuYoIMK8MIp/PaINU0wWSKx9S3VGJBdZAVt87RmVVGKIqVLWlQof6eyLDQeiZC2ymwmehlLxf/8/zURNdBxmSSGirJYlGUcmRilD+ORkxRYvjMEkwUs7ciMsEKE2PjqdkQvOWXV0nvsum1mp5312q0L8o8qnACp3AOHlxBG26hA10gMIFneIU3RzgvzrvzsWitOOXMMfyB8/kDpdiOYw==</latexit>

x[n]
A/D

fs = 240Hz
<latexit sha1_base64="ONUaok1RM9k3HiC2ePuggdtN6NU=">AAACB3icbVBNS8NAEN34WetX1KMIi0XwICUpBb0IBS89VrC20Iaw2W7apZsPdidiDTl58a948aAgXv0L3vw3btoctPXBwOO9GWbmebHgCizr21haXlldWy9tlDe3tnd2zb39WxUlkrI2jUQkux5RTPCQtYGDYN1YMhJ4gnW88VXud+6YVDwKb2ASMycgw5D7nBLQkmse+W7aDwiMZJCqLMOXuFa3+sDuIW0+ZK5ZsarWFHiR2AWpoAIt1/zqDyKaBCwEKohSPduKwUmJBE4Fy8r9RLGY0DEZsp6mIQmYctLpGxk+0coA+5HUFQKeqr8nUhIoNQk83ZlfrOa9XPzP6yXgXzgpD+MEWEhni/xEYIhwngkecMkoiIkmhEqub8V0RCShoJMr6xDs+ZcXSadWtetV276uVxpnRR4ldIiO0Smy0TlqoCZqoTai6BE9o1f0ZjwZL8a78TFrXTKKmQP0B8bnD5FQmOs=</latexit>

• Problem: x0 (t) is contaminated with fi = 60Hz interference

<latexit sha1_base64="UVBLAzJ4yI7sdZGte0OA5y3cWMo=">AAAB7XicbVBNS8NAEJ3Ur1q/qh69BItYLyUrBT0WvHisYGyhDWWz3bRLN5uwOxFL6W/w4kFBvPp/vPlv3LY5aOuDgcd7M8zMC1MpDHret1NYW9/Y3Cpul3Z29/YPyodHDybJNOM+S2Si2yE1XArFfRQoeTvVnMah5K1wdDPzW49cG5GoexynPIjpQIlIMIpW8p/Oq3jRK1e8mjeHu0pITiqQo9krf3X7CctirpBJakyHeCkGE6pRMMmnpW5meErZiA54x1JFY26CyfzYqXtmlb4bJdqWQneu/p6Y0NiYcRzazpji0Cx7M/E/r5NhdB1MhEoz5IotFkWZdDFxZ5+7faE5Qzm2hDIt7K0uG1JNGdp8SjYEsvzyKmld1ki9RshdvdIgeR5FOIFTqAKBK2jALTTBBwYCnuEV3hzlvDjvzseiteDkM8fwB87nD3Svjjg=</latexit>

<latexit sha1_base64="6FVBieVJVL/0aE8u+DijeTzOw3w=">AAACB3icbVBNS8NAEN3Ur1q/oh5FWCyCp5JIUS9CwUuPFawttCFstpt26eaD3YlYQ05e/CtePCiIV/+CN/+NmzYHbX0w8Hhvhpl5Xiy4Asv6NkpLyyura+X1ysbm1vaOubt3q6JEUtamkYhk1yOKCR6yNnAQrBtLRgJPsI43vsr9zh2TikfhDUxi5gRkGHKfUwJacs1D3037AYGRDFKeZfgSn1m4D+we0uZD5ppVq2ZNgReJXZAqKtByza/+IKJJwEKggijVs60YnJRI4FSwrNJPFIsJHZMh62kakoApJ52+keFjrQywH0ldIeCp+nsiJYFSk8DTnfnFat7Lxf+8XgL+hZPyME6AhXS2yE8EhgjnmeABl4yCmGhCqOT6VkxHRBIKOrmKDsGef3mRdE5rdr1m29f1asMu8iijA3SETpCNzlEDNVELtRFFj+gZvaI348l4Md6Nj1lryShm9tEfGJ8/ahmY1g==</latexit>

cycles
<latexit sha1_base64="BfMjC9Rap96ejLt6BqEFNaGzgPM=">AAACM3icbVDLSgNBEJz1bXxFPXoZDIKnsCuKXoSAF70pGBPIhjA76U2GzM4sM71iXPJNXvwSwYMeFMSr/+DkcfCRgoaiqpvuriiVwqLvv3gzs3PzC4tLy4WV1bX1jeLm1o3VmeFQ5VpqU4+YBSkUVFGghHpqgCWRhFrUOxv6tVswVmh1jf0UmgnrKBELztBJreJFTE+pXz44omFsGM9DhDvMw1hrVBrBinugvM8l2MFgqmlZko7cVrHkl/0R6H8STEiJTHDZKj6Fbc2zBBRyyaxtBH6KzZwZFG7foBBmFlLGe6wDDUcVS8A289HLA7rnlDaNtXGlkI7UnxM5S6ztJ5HrTBh27V9vKE7zGhnGJ81cqDRDUHy8KM4kRU2H+dG2MMBR9h1h3Ah3K+Vd5qJDl3LBhRD8ffk/qR2Ug8NyEFwdlirBJI8lskN2yT4JyDGpkHNySaqEkwfyTN7Iu/fovXof3ue4dcabzGyTX/C+vgEB7axR</latexit>

• For fs = 240Hz , 60Hz interference is sampled 4 times per cycle, i.e., f = 0.25
<latexit sha1_base64="q/lBl7iDJ5WjHBHik+lvuK2cZOI=">AAACCHicbVBNS8NAEN3Ur1q/oh4FWSyCp5KUgl6EgpceK1hbaErYbDft0s0HuxOxhty8+Fe8eFAQr/4Eb/4bN20O2vpg4PHeDDPzvFhwBZb1bZRWVtfWN8qbla3tnd09c//gVkWJpKxDIxHJnkcUEzxkHeAgWC+WjASeYF1vcpX73TsmFY/CG5jGbBCQUch9TgloyTWPfTd1AgJjGaQqy/Alrjcs7AC7h7T1kLlm1apZM+BlYhekigq0XfPLGUY0CVgIVBCl+rYVwyAlEjgVLKs4iWIxoRMyYn1NQxIwNUhnf2T4VCtD7EdSVwh4pv6eSEmg1DTwdGd+slr0cvE/r5+AfzFIeRgnwEI6X+QnAkOE81DwkEtGQUw1IVRyfSumYyIJBR1dRYdgL768TLr1mt2o2fZ1o9q0izzK6AidoDNko3PURC3URh1E0SN6Rq/ozXgyXox342PeWjKKmUP0B8bnD/DtmRo=</latexit>

<latexit sha1_base64="3V7Ct2ZOOrt7+RVn3D0YeQ2SRds=">AAAB9HicbVDLSgNBEOyNrxhfUY9eBoPgKexIUI8BLzlGMCaQLGF2MpsMmX0w0yvGJb/hxYOCePVjvPk3TpI9aGJBQ1HVTXeXnyhp0HW/ncLa+sbmVnG7tLO7t39QPjy6N3GquWjxWMW64zMjlIxECyUq0Um0YKGvRNsf38z89oPQRsbRHU4S4YVsGMlAcoZW6l26pIfiEbPG07RfrrhVdw6ySmhOKpCj2S9/9QYxT0MRIVfMmC51E/QyplFyJaalXmpEwviYDUXX0oiFwnjZ/OYpObPKgASxthUhmau/JzIWGjMJfdsZMhyZZW8m/ud1UwyuvUxGSYoi4otFQaoIxmQWABlILTiqiSWMa2lvJXzENONoYyrZEOjyy6ukfVGltSqlt7VKneZ5FOEETuEcKFxBHRrQhBZwSOAZXuHNSZ0X5935WLQWnHzmGP7A+fwBE2WRew==</latexit>

<latexit sha1_base64="v4vGRtl1Th/dRT/CHQImjQqoIHY=">AAAB5XicbVBNS8NAEJ34WeNX9eplsQieSlYKeix48VjB2kIbymY7adduNmF3I5TQX+DFg4JX/5I3/43bNgdtfTDweG+GmXlRJoWxQfDtbWxube/sVvb8/YPDo+Oqf/Jo0lxzbPNUprobMYNSKGxbYSV2M40siSR2osnt3O88ozYiVQ92mmGYsJESseDMOum+MajWgnqwAFkntCQ1KNEaVL/6w5TnCSrLJTOmR4PMhgXTVnCJM7+fG8wYn7AR9hxVLEETFotDZ+TCKUMSp9qVsmSh/p4oWGLMNIlcZ8Ls2Kx6c/E/r5fb+CYshMpyi4ovF8W5JDYl86/JUGjkVk4dYVwLdyvhY6YZty4b34VAV19eJ52rOm3UKa01aZlGBc7gHC6BwjU04Q5a0AYOCC/wBu/ek/fqfSwbN7xy4hT+wPv8AaSii7U=</latexit>

samples

<latexit sha1_base64="itGfDU4ktaJx2j5fPVShyuqpmk8=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECvVY8OKxgrWFNJTNdtMs3d2E3Y1QQv+CFw8K4tUf5M1/46bNQVsfDDzem2FmXphypo3rfjuVjc2t7Z3qbm1v/+DwqH588qiTTBHaIwlP1CDEmnImac8ww+kgVRSLkNN+OL0t/P4TVZol8sHMUhoIPJEsYgSbQop9GYzqDbfpLoDWiVeSBpTojupfw3FCMkGlIRxr7XtuaoIcK8MIp/PaMNM0xWSKJ9S3VGJBdZAvbp2jC6uMUZQoW9Kghfp7IsdC65kIbafAJtarXiH+5/mZiW6CnMk0M1SS5aIo48gkqHgcjZmixPCZJZgoZm9FJMYKE2PjqdkQvNWX10n/uum1mp5332p0rso8qnAG53AJHrShA3fQhR4QiOEZXuHNEc6L8+58LFsrTjlzCn/gfP4AjViOUw==</latexit>

h[n]
<latexit sha1_base64="teIAQSHnq3BmnosM5F+tbG+jXjA=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECnosePFYwdpCGspmu2mX7m7C7kYsoX/BiwcF8eoP8ua/cZPmoK0PBh7vzTAzL0w408Z1v53K2vrG5lZ1u7azu7d/UD88etBxqgjtkpjHqh9iTTmTtGuY4bSfKIpFyGkvnN7kfu+RKs1ieW9mCQ0EHksWMYJNLj35MhjWG27TLYBWiVeSBpToDOtfg1FMUkGlIRxr7XtuYoIMK8MIp/PaINU0wWSKx9S3VGJBdZAVt87RmVVGKIqVLWlQof6eyLDQeiZC2ymwmehlLxf/8/zURNdBxmSSGirJYlGUcmRilD+ORkxRYvjMEkwUs7ciMsEKE2PjqdkQvOWXV0nvsum1mp5312q0L8o8qnACp3AOHlxBG26hA10gMIFneIU3RzgvzrvzsWitOOXMMfyB8/kDpdiOYw==</latexit>

<latexit sha1_base64="RMhumvSiRLzep4axThwabeR3QqA=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECvVY8OKxgrWFNJTNdtMu3d2E3Y0QQv+CFw8K4tUf5M1/46bNQVsfDDzem2FmXphwpo3rfjuVjc2t7Z3qbm1v/+DwqH588qjjVBHaIzGP1SDEmnImac8ww+kgURSLkNN+OLst/P4TVZrF8sFkCQ0EnkgWMYJNIWW+DEb1htt0F0DrxCtJA0p0R/Wv4TgmqaDSEI619j03MUGOlWGE03ltmGqaYDLDE+pbKrGgOsgXt87RhVXGKIqVLWnQQv09kWOhdSZC2ymwmepVrxD/8/zURDdBzmSSGirJclGUcmRiVDyOxkxRYnhmCSaK2VsRmWKFibHx1GwI3urL66R/3fRaTc+7bzU6V2UeVTiDc7gED9rQgTvoQg8ITOEZXuHNEc6L8+58LFsrTjlzCn/gfP4Ap2COZA==</latexit>

x[n] y[n]
Filter

place zeros on
<latexit sha1_base64="JrPVjZTMjLk+0nc4y+J4VkoH8K4=">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</latexit>

fi 60 ⇡ rad
!i = 2⇡ = 2⇡ =
fs 240 2 samples interference
1

1
 10/17/24

Frequency Domain Representation

<latexit sha1_base64="itGfDU4ktaJx2j5fPVShyuqpmk8=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECvVY8OKxgrWFNJTNdtMs3d2E3Y1QQv+CFw8K4tUf5M1/46bNQVsfDDzem2FmXphypo3rfjuVjc2t7Z3qbm1v/+DwqH588qiTTBHaIwlP1CDEmnImac8ww+kgVRSLkNN+OL0t/P4TVZol8sHMUhoIPJEsYgSbQop9GYzqDbfpLoDWiVeSBpTojupfw3FCMkGlIRxr7XtuaoIcK8MIp/PaMNM0xWSKJ9S3VGJBdZAvbp2jC6uMUZQoW9Kghfp7IsdC65kIbafAJtarXiH+5/mZiW6CnMk0M1SS5aIo48gkqHgcjZmixPCZJZgoZm9FJMYKE2PjqdkQvNWX10n/uum1mp5332p0rso8qnAG53AJHrShA3fQhR4QiOEZXuHNEc6L8+58LFsrTjlzCn/gfP4AjViOUw==</latexit>

h[n]
<latexit sha1_base64="teIAQSHnq3BmnosM5F+tbG+jXjA=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECnosePFYwdpCGspmu2mX7m7C7kYsoX/BiwcF8eoP8ua/cZPmoK0PBh7vzTAzL0w408Z1v53K2vrG5lZ1u7azu7d/UD88etBxqgjtkpjHqh9iTTmTtGuY4bSfKIpFyGkvnN7kfu+RKs1ieW9mCQ0EHksWMYJNLj35MhjWG27TLYBWiVeSBpToDOtfg1FMUkGlIRxr7XtuYoIMK8MIp/PaINU0wWSKx9S3VGJBdZAVt87RmVVGKIqVLWlQof6eyLDQeiZC2ymwmehlLxf/8/zURNdBxmSSGirJYlGUcmRilD+ORkxRYvjMEkwUs7ciMsEKE2PjqdkQvOWXV0nvsum1mp5312q0L8o8qnACp3AOHlxBG26hA10gMIFneIU3RzgvzrvzsWitOOXMMfyB8/kDpdiOYw==</latexit>

<latexit sha1_base64="RMhumvSiRLzep4axThwabeR3QqA=">AAAB7HicbVBNS8NAEJ3Ur1q/qh69LBbBg5RECvVY8OKxgrWFNJTNdtMu3d2E3Y0QQv+CFw8K4tUf5M1/46bNQVsfDDzem2FmXphwpo3rfjuVjc2t7Z3qbm1v/+DwqH588qjjVBHaIzGP1SDEmnImac8ww+kgURSLkNN+OLst/P4TVZrF8sFkCQ0EnkgWMYJNIWW+DEb1htt0F0DrxCtJA0p0R/Wv4TgmqaDSEI619j03MUGOlWGE03ltmGqaYDLDE+pbKrGgOsgXt87RhVXGKIqVLWnQQv09kWOhdSZC2ymwmepVrxD/8/zURDdBzmSSGirJclGUcmRiVDyOxkxRYnhmCSaK2VsRmWKFibHx1GwI3urL66R/3fRaTc+7bzU6V2UeVTiDc7gED9rQgTvoQg8ITOEZXuHNEc6L8+58LFsrTjlzCn/gfP4Ap2COZA==</latexit>

x[n] y[n]
Filter

zero locations to cancel

<latexit sha1_base64="enmBR+PbidwhfydHS16FRDEXMdQ=">AAACE3icbZBNS8NAEIY39avWr6hHL4tFUMSSLQW9CAUvHitYW0hD2Gw37dLNJuxuxFD6G7z4V7x4UBCvXrz5b9ykPWjrwMIz78wwO2+QcKa043xbpaXlldW18nplY3Nre8fe3btTcSoJbZOYx7IbYEU5E7Stmea0m0iKo4DTTjC6yuudeyoVi8WtzhLqRXggWMgI1kby7ZPMFR68hIHvwIccTw2iHM/QNKkXSd3z7apTc4qAi4BmUAWzaPn2V68fkzSiQhOOlXKRk2hvjKVmhNNJpZcqmmAywgPqGhQ4osobFydN4JFR+jCMpXlCw0L9PTHGkVJZFJjOCOuhmq/l4n81N9XhhTdmIkk1FWS6KEw51DHM/YF9JinRPDOAiWTmr5AMscREGxcrxgQ0f/IidOo11KghdNOoNtHMjzI4AIfgGCBwDprgGrRAGxDwCJ7BK3iznqwX6936mLaWrNnMPvgT1ucP0T+aLA==</latexit>
interference poles
y[n] = b0 x[n] + b1 x[n 1] + b2 x[n 2]
<latexit sha1_base64="5vPNhZZiuR8jc6JYfJhtsNDhIbY=">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</latexit>

p
<latexit sha1_base64="DRJktJ/5wBWCSbgLqsMwMW2RvoQ=">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</latexit>

1 2 b1 ± b21 4b0 b2
Y (z) = b0 X(z) + b1 z X(z) + b2 z X(z) z1 , z2 =
2b0
1 2
= (bo + b1 z + b2 z )X(z)

b 0 z 2 + b 1 z + b2
<latexit sha1_base64="SKQScNWm3LBKWUAvRrCwQmEnVkc=">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</latexit>

Y (z) 1 2 (z z1 )(z z2 )
H(z) = = b 0 + b1 z + b2 z = 2
=
X(z) z z2
2

Z-Transform Examples
z3 + z2 + z + 1
<latexit sha1_base64="xLzgO9Owj+3nAWVARpKF1XZcoMc=">AAACEnicbVDLSgMxFM3UV62vUZdugkWpCGVSC7oRCm66rGBtoZ2WTJppQzMPkozQDvMLbvwVNy4UxK0rd/6NmXYW2nrgksM593JzjxNyJpVlfRu5ldW19Y38ZmFre2d3z9w/uJdBJAhtkoAHou1gSTnzaVMxxWk7FBR7DqctZ3yT+q0HKiQL/Ds1Cant4aHPXEaw0lLfLNVL0zN4eg27rsAknvYu4Dmc9uJKkr66UJKKCeybRatszQCXCcpIEWRo9M2v7iAgkUd9RTiWsoOsUNkxFooRTpNCN5I0xGSMh7SjqY89Ku14dlECT7QygG4gdPkKztTfEzH2pJx4ju70sBrJRS8V//M6kXKv7Jj5YaSoT+aL3IhDFcA0HjhgghLFJ5pgIpj+KyQjrKNROsSCDgEtnrxMWpUyqpYRuq0Wa1aWRx4cgWNQAghcghqogwZoAgIewTN4BW/Gk/FivBsf89ackc0cgj8wPn8An9SaJA==</latexit>

H(z) =
z3
<latexit sha1_base64="q6k9Gs1a7KjwtVpu7H4bJCzkARQ=">AAACAXicbVBNS8NAEN3Ur1q/oh48eAkWoYKURAp6LHjxWMHaQhvKZjtpl24+3J2INeTiX/HiQUG8+i+8+W/ctjlo6xsGHu/NsDvPiwVXaNvfRmFpeWV1rbhe2tjc2t4xd/duVZRIBk0WiUi2PapA8BCayFFAO5ZAA09AyxtdTvzWPUjFo/AGxzG4AR2E3OeMopZ65kEX4QFTX8LdY1bpOKfTck96Ztmu2lNYi8TJSZnkaPTMr24/YkkAITJBleo4doxuSiVyJiArdRMFMWUjOoCOpiENQLnp9IDMOtZK3/IjqTtEa6r+3khpoNQ48PRkQHGo5r2J+J/XSdC/cFMexglCyGYP+YmwMLImaVh9LoGhGGtCmeT6rxYbUkkZ6sxKOgRn/uRF0jqrOrWq41zXynU7z6NIDskRqRCHnJM6uSIN0iSMZOSZvJI348l4Md6Nj9lowch39skfGJ8/ShWVSg==</latexit> <latexit sha1_base64="QK7GlbciB+qE1WWwTZ1LrJpjnU4=">AAACAnicbVBNS8NAEN3Ur1q/ql4EL8EiVJCSSEGPBS8eK1hbSEPZbCft0s0m7E7EGurFv+LFg4J49Vd489+4/Tho6xsGHu/NsDsvSATX6DjfVm5peWV1Lb9e2Njc2t4p7u7d6jhVDBosFrFqBVSD4BIayFFAK1FAo0BAMxhcjv3mHSjNY3mDwwT8iPYkDzmjaKRO8aCNcI/ZQyKohFHZc08n5Z90iiWn4kxgLxJ3Rkpkhnqn+NXuxiyNQCITVGvPdRL0M6qQMwGjQjvVkFA2oD3wDJU0Au1nkwtG9rFRunYYK9MS7Yn6eyOjkdbDKDCTEcW+nvfG4n+el2J44WdcJimCZNOHwlTYGNvjOOwuV8BQDA2hTHHzV5v1qaIMTWgFE4I7f/IiaZ5V3GrFda+rpZozyyNPDskRKROXnJMauSJ10iCMPJJn8krerCfrxXq3PqajOWu2s0/+wPr8AQ3flbY=</latexit>

freqz([1, 1, 1, 1]) zplane([1, 1, 1, 1])

2
 10/17/24

Z-Transform Examples
z3 + z2 + z + 1
<latexit sha1_base64="twyX6ISabw6Z2Z2JDxE1jeJhsHs=">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</latexit>

H(z) =
z 3 + 0.9z 2 + 0.81z + 0.729
<latexit sha1_base64="SIeWCtS63Q9Ds//bW4unZKuUI1M=">AAACFHicbZA7SwNBEMf3fMb4ilraHAYhSgh3IZCkC9hYRjAmkBxhbzOXLNl7uDsnxiPfwcavYmOhILYWdn4bN49CE2cZ+PGfGWbn70aCK7Ssb2NldW19YzO1ld7e2d3bzxwc3qgwlgwaLBShbLlUgeABNJCjgFYkgfqugKY7vJjUm3cgFQ+DaxxF4Pi0H3CPM4pa6mbOOwj3mHgSbh/Gubadnz4nr8kqVHVWJlAuVp2zbiZrFaxpmMtgzyFL5lHvZr46vZDFPgTIBFWqbVsROgmVyJmAcboTK4goG9I+tDUG1AflJNObxuapVnqmF0qdAZpT9fdEQn2lRr6rO32KA7VYm4j/1doxehUn4UEUIwRstsiLhYmhOTHI7HEJDMVIA2WS67+abEAlZahtTGsT7MWTl6FZLNilgm1flbI1a+5HihyTE5IjNimTGrkkddIgjDySZ/JK3own48V4Nz5mrSvGfOaI/Anj8wfIa5n/</latexit> <latexit sha1_base64="Tw2AAcgkejrGeWIwXIMJzvdI+S0=">AAACFXicbZBNS8NAEIY3ftb6VfXoJViEiiUkIrTeBC8eK1grtKFsttN2cbMJuxOxhv4IL/4VLx4UxKvgzX/jJubg1ywDD+/MMDtvEAuu0XU/rJnZufmFxdJSeXlldW29srF5oaNEMWizSETqMqAaBJfQRo4CLmMFNAwEdIKrk6zeuQaleSTPcRKDH9KR5EPOKBqpX9nvIdxgehsLKmFa63r1/Pl1Q65zZLKZQePgyN/rV6qu4+Zh/wWvgCopotWvvPcGEUtCkMgE1brruTH6KVXImYBpuZdoiCm7oiPoGpQ0BO2n+VFTe9coA3sYKZMS7Vz9PpHSUOtJGJjOkOJY/65l4n+1boLDpp9yGScIkn0tGibCxsjOHLIHXAFDMTFAmeLmrzYbU0UZGh/LxgTv98l/oXPgeIeO550dVo/dwo8S2SY7pEY80iDH5JS0SJswckceyBN5tu6tR+vFev1qnbGKmS3yI6y3T5Ohmms=</latexit>

freqz([1, 1, 1, 1], [1, 0.9, 0.81, 0.729]) zplane([1, 1, 1, 1], [1, 0.9, 0.81, 0.729])

Z-Plane Pole Locations and Their Corresponding Time Series

3
 10/17/24

Z-Plane Pole Locations and Their Corresponding Time Series

• Imp

4
Filter Design: Low Pass Filter Design

Problem: Given δ1 , δ 2 , ω p and ωs , find the lowest complexity filter

that meets specification
Two Choices
B( z )
1. IIR Filters (Infinite Impulse Response, H ( z) =
A( z )
2. FIR Filters (Finite Impulse Response, H ( z ) = B( z )
IIR Filter Design
 Approach: Transform a continuous time filter
to a discrete time filter

Rationale:
 Continuous time IIR filter design highly
advanced.
 Many continuous time IIR filters have closed
form design formulas
 Approximating directly in the digital domain
not easy.
 IIR Filter Design Methods

1. Impulse Invariance
2. Bilinear Transformation

Another useful Tool: Frequency Transformation.

It allows design of other filter types (High
Pass, Band Pass…) by transforming a low pass
filter.
Impulse Invariance Design
Simply applies the sampling theorem and
hopes that aliasing is small
h[n] = Td hc (nTd )
Then ∞
ω
2π
H (e ) = ∑ H c ( j
jω
− j k)
k = −∞ Td Td
More generally
∞
2π
H ( z) z = e sTd
= ∑ H c (s − j k)
k = −∞ Td
 N
Ak
H c ( s) = ∑
k =1 s − sk
Then
N
hc (t ) = ∑ k , t≥0
A e
k =1
sk t

N
h[ n] = Td hc ( nTd ) = Td ∑ Ak e sk Td n , n ≥ 0
k =1
N
Td Ak
H ( z) = ∑
k =1 1 − e
s k Td
z −1

Pole of H c (s ) at sk gets transformed to a pole of H(z) at e sk Td

Stability is preserved
Problem: How to account for aliasing
Similar approach can be used for step invariance or other waveform
invariance criteria
 Impulse Variance Design
(Cont.)
Example

s + .1 .5 .5
H c ( s) = = +
( s + .1) + 9 s + .1 − j 3 s + .1 + j 3
2

.5Td .5Td
H ( z) = −.1Td j 3Td −1
+ −.1Td − j 3Td −1
1− e e z 1− e e z
Analog
Filter

Impulse Invariance
Design
 Bi-Linear Transformation
2 1 − z −1
s= −1
Td 1 + z
The digital filter is obtained by a change of variables

 2 1 − z −1 
H ( z) = H c  −1 
 d
T 1 + z 
σ Ω
2
σTd  d 
2 2
Td ΩTd T Td
1+ s 1+ +j 1 +  +
or z =  
2 = 2 2 2 4
z=
2

Td
1− s 1− σT Ω T  σTd 
2
Ω 2 2
Td
d
−j d
1 −  +
2 2 2  2  4
Bilinear Transformation σTd  Ω
2

2 2
Td
1 +  +
z =
2  2  4
Properties of Mapping
σ Ω
2
 d 
2 2
T Td
1 −  +
σ<0 |z| < 1  2  4
σ=0 |z| = 1 σ>0 |z| > 1
Mapping and its Role in Filter Design
ΩTd
1+ j 2 ω   ΩTd 
e jω = 2 or Ω = tan   or ω = 2 arctan  
ΩTd Td 2  2 
1− j
2

Mapping between the frequency ω and Ω resulting from the

Bilinear transformation
Frequency warping inherent in the bi-linear transformation of a
continuous-time lowpass filter into a discrete-time lowpass filter.
Illustration of the effect of the bi-linear transformation on a linear phase
characteristic
Overall Design Steps
Given δ1 , δ 2 , ωp , ωs
1. Pre-warp to get δ1 , δ 2 , Ωp , Ω s
2  ωi 
where Ωi = tan  
Td  2 
2. Using these specs design Analog
Filter H ( jΩ) or H ( s )
c c

 2 1− z  −1
3. H ( z ) = H c 
−1 
 Td 1 + z 
Butterworth Filters as a Type I Chebyshev
Prof Rao
function of filter order N Lowpass Filter

Elliptic lowpass filter

Overall Design Steps
Given δ1 , δ 2 , ωp , ωs
1. Pre-warp to get δ1 , δ 2 , Ωp , Ω s
2  ωi 
where Ωi = tan  
Td  2 
2. Using these specs design Analog
Filter H ( jΩ) or H ( s )
c c

 2 1− z  −1
3. H ( z ) = H c 
−1 
 Td 1 + z 
Butterworth Filters as a Type I Chebyshev
Prof Rao
function of filter order N Lowpass Filter

Elliptic lowpass filter

Butterworth Filter

Real coefficient filters

1
H ( jΩ) =
2
2N
 Ω 
1+ 
Ω 

 c 
1 ( jΩ c ) 2 N
H ( s) H (− s) = = 2N
 s 
2N
s + ( jΩ c ) 2 N
1+ 
 jΩ 

 c 
Roots of the Butterworth Filter

s 2N + (jΩc )2N = 0 or s 2N = −(jΩc )2N

Now −1 = e j(2k−1)π , k = 0, 1, .. allowing us to compute the 2N roots as
follows

sk2N = e j(2k−1)π (jΩc )2N

(2k−1)π (2k−1)π π
sk = ej 2N jΩc = e j 2N e j 2 Ωc
2kπ π π π π 2kπ
= ej 2N e j( 2 − 2N ) Ωc = e j( 2 − 2N ) e j 2N Ωc
Butterworth Filter Cont’d
N=3

Poles of filter are on a circle of radius Ω c

π π j 2π k jπ
j( − ) (2 k + N −1)
sk =
Ωc e 2 2N
e 2N
=
Ωc e ,k =
2N
0,1,..., 2 N − 1
 Butterworth Filters (cont.)
H c (s ) is formed by taking all the poles in the
left half plane
(Ω c ) N
H c ( s) =
∏k =1 ( s − sk )
N
Design Example
 Specification of Digital Filter
.89125 ≤ H ( e jω ) ≤ 1, 0 ≤ ω ≤ .2π

H ( e jω ) ≤ 0.17783, .3π ≤ ω ≤ π
 Specifications of Analog Filter
2  0.2π 
.89125 ≤ H ( jΩ) ≤ 1, 0 ≤ Ω ≤ tan  
Td  2 
2  0.3π 
H ( jΩ) ≤ 0.17783, tan  ≤Ω≤∞
Td  2 

 Design Constraints Td = 1
H c ( j 2 tan(.1π ) ≥ .89
H c ( j 2 tan(.15π ) ≤ .178
Design Example (cont.)
1
For the Butterworth filter H ( jΩ ) = and
2
 2N
Ω
due to its monotonic response 1 +  
 Ωc 
2N
 2 tan(.1π ) 
2
 1 
1+ 
 
 = 
 Ω c   .89 
2N
 2 tan(.15π ) 
2
 1 
1+ 
 
 = 
 Ω c   .178 

 Solving for N, yields N= 5.3. Since N has to

be an integer it is rounded up and N=6.
Solving for Ω c from the stop-band specs
yields Ω c =.76622
 Design Example (Cont. Pg 2)
N =6
Ω c = .766

S-Plane locations for pole of H c ( s) H c (− s) for sixth-

order Butterworth Filter in Example 7.4
 Design Examples (Cont. Pg 3)

0.20238
H c ( s) = 2
( s + .3996s + 0.59)( s 2 + 1.084 s + 0.59)( s 2 + 1.48s + 0.59)
 1 − z −1 
H ( z ) = H c  2 
−1 
 1+ z 
0.0007378(1 + z −1 )6
=
(1 − 1.27 z −1 + .71z −2 )(1 − 1.01z −1 + .36 z −2 )(1 − .9 z −1 + .22 z −2 )
Bandpass Filter Example

It Ew
i181

o Us Wp
M WpsWs2 W IT

Digital Filter Specifications: δ1 , δ2 , δ3 , ωs1 , ωp1 , ωp2 , ωs2

Analog Filter Specifications

Digital Filter Specifications: δ1 , δ2 , δ3 , ωs1 , ωp1 , ωp2 , ωs2

The δ’s remain the same. Pre-Warp to get the Analog Filter band edges.

2 ωs1
Ωs 1 = tan
Td 2
2 ωp1
Ωp 1 = tan
Td 2
2 ωp2
Ωp 2 = tan
Td 2
2 ωs2
Ωs 2 = tan
Td 2
Analog Filter Specifications: δ1 , δ2 , δ3 , Ωs1 , Ωp1 , Ωp2 , Ωs2
Frequency Transformation
 Change of variable z −1 → G ( z −1 ) to get desired
response (low pass, high pass, band pass,
etc.)
H ( z) 
→ H d ( z ) = H ( z ) z −1 → G ( z −1 )
 Alternate Notation
Z −1 = G ( z −1 )
H d ( z ) = H ( Z ) Z −1 =G ( z −1 )
 Constraints on G(.)
 Stability be preserved.
 G(.) is a rational function
Lowpass to Lowpass
z −1
−α
Z −1 = , α real
1 − αz −1

Also
Z −1 + α
z −1
= , α real
1 + αZ −1

If Z = e jθ and z = e jω , then
e − jθ
+α
e − jω =
1 + αe − jθ
 (1 − α 2 ) sin θ 
ω = arctan  

 2α + (1 + α ) cos θ
2

θ p − ωp
sin
α = 2
θ p + ωp
sin
2
Lowpass to Highpass
z +α
−1
Z −1
=− Choose α as
1 + αz −1
follows

θ p + ωp
cos
α =− 2
θ p − ωp
cos
2
α = 0→ ω p = π − θ p