Lecture #1

Review of DT Signal
& System Concepts

1
Notational Conventions (P-1.2)
Sets of Numbers: R = real numbers
C = complex numbers
Z = integers ( …, -3, -2, -1, 0, 1, 2, 3, …)

Signals: x(t), y(t) = continuous-time (CT) signals

x[n], y[n] = discrete-time (DT) signals

Modulo Operator – “Clock Math”: n mod N

n: … -4 -3 -2 -1 0 1 2 3 4 5 6 7 … “15 o’clock” = 15 mod 12
n mod N: … 2 0 1 2 0 1 2 0 1 2 0 1 … = “3 o’clock”
∞
DT Convolution: {x * y}[n ] = ∑ x[ m ] y [ n − m ]
m = −∞

N −1
DT Circular Convolution: {x ⊕
⊗ y}[n ] = ∑ x[m] y[(n − m) mod N ] 0 ≤ n ≤ N −1
m =0

Recall: We don’t want Circular convolution – it is what we get if we try to implement

Convolution in the frequency domain by multiplying DFT’s (but not DTFT’s)
But… we can fix it so that it works…. We use proper zero-padding!!! 2
Transforms (P-2)
Fourier Transform for CT Signals Fourier Transform for DT Signals
X F (ω ) = {Fx}(ω ) X f (θ ) = {Fx}(θ ) Discrete
∞ ∞
Time
= ∫ x (t )e
− jωt
dt = ∑ x[n]e − jθn Fourier
−∞ n = −∞
Transform
∞ π
1
1 jωt
∫X (θ )e jθn dθ
∫ X (ω )e dω x[ n ] =
F f
x (t ) =
2π 2π
−∞ −π

Set z = ejθ

Z Transform for DT Signals Discrete Fourier Transform for DT Signals

X z ( z ) = {Zx}( z ) Discrete Fourier Transform
∞ X [k ] = {Dx}[k ]
d
= ∑ x[n]z −n
N −1
n = −∞ = ∑ x[n]e − j 2πkn / N k = 0,1,2,..., N − 1
n =0
Inverse ZT done using partial
fractions & a ZT table N −1
∑ X d [k ]e j 2πkn / N
1
x[n ] = n = 0,1,2,..., N − 1
N k =0 3
Fourier Transforms for DT Signals (P-2.7,P-4)
DTFT = “In Head/On Paper” tool for analysis & design

DFT = “In Computer” tool for Implementation/Analysis&Design

(In both cases you generally strive to make the DFT behave like DTFT)

• DFT for Implementation:

! Compute DFT of signal to be processed and manipulate DFT values
! May or may not need zero-padding (depends on application)
• DFT for Analysis & Design:
! Compute DFT of given filter’s TF to see its frequency response
! Generally use zero-padding to get fine grid to get approximate DTFT

DFT/DTFT Properties you Must Know:

• Periodicity…………………. (DFT&DTFT both periodic in FD)
…………………. (IDFT gives periodic TD signal)
• Time Shift………………….. (Time Shift in TD " Linear Phase Shift in FD)
• Frequency Shift……………. (Mult. by complex sine in TD " Freq Shift in FD)
• Convolution Theorem ……… (Conv. in Time Domain " Mult. in Freq Domain)
• Multiplication Theorem…….. (Mult. in Time Domain " Conv. in Freq Domain)
• Parseval’s Theorem…………. (Sum of Squares in TD = Sum of Squares in FD)
4
Summation Rules (P-1.3)
Make sure you are familiar with rules for dealing with
summations, as shown in Section 1.3 of Porat.
In addition, a particular summation formula that shows up
often is the “Geometric Summation”:
 α N1 − α N 2
 , if α ≠ 1
N 2 −1  1−α
n 
∑ α =
n = N1 N − N , if α = 1
 2 1


Special Case: N1 = 0
1 − α N 2
 , if α ≠ 1
N 2 −1  1 − α
∑ αn = 
n =0 N , if α = 1
 2
 5
Euler’s Formulas
e jθ + e − jθ
cos(θ ) =
2
jθ e jθ − e − jθ
e = cos(θ ) + j sin(θ ) sin(θ ) =
2j
e − jθ = cos(θ ) − j sin(θ )
Im
Im

e jθ e jθ

2jsin(θ)
− jθ
−e
j sin(θ )

cos(θ ) Re 2 cos(θ ) Re
− j sin(θ )

e − jθ e − jθ

6
Discrete-Time Processing (P-2, P-7)
x[n] h[n] y[n] = h[n]*x[n]
Xf(θ) Hf(θ) Yf(θ) = Hf(θ) Xf(θ)
Xz(z) Hz(z) Yz(z) = Hz(z) Xz(z)

h[n] = system’s “Impulse Response”

It is response of system to input of δ[n]

Hf(θ) = system’s “Frequency Response”

It is DTFT of impulse response

Hz(z) = system’s “Transfer Function”

It is ZT of impulse response

7
 Discrete-Time System Relationships
Time Domain Z / Freq Domain

Pole/Zero
Block Diagram
Inspect Diagram
Inspect Roots

Difference ZT (Theory) Transfer z b0 + b1 z −1 + L + bq z − q

H ( z) =
Equation Inspect (Practice) Function 1 + a1 z −1 + L + a p z − p
p q
y[k ] = − ∑ ai y[n − i ] + ∑ bi x[n − i ]
i =1 i =0
Unit Circle
ZT
z = ejθ

Frequency
DTFT
Impulse Response
Response
H f (θ ) = H z ( z ) |z = e jθ
8
h[k]
 Poles and Zeros of Transfer Function

Consider Here Only p>q

Im{z}
−1 −q
b0 + b1 z + L + bq z
H z ( z) =
1 + a1 z −1 + L + a p z − p
z p − q z q (b0 + b1 z −1 + L + bq z − q )
=
z p (1 + a1 z −1 + L + a p z − p )
z p − q (b0 z q + b1 z q −1 + L + bq ) Re{z}
=
z p + a1 z p −1 + L + a p
z p − q ( z − β1 )( z − β 2 )L( z − β q )
=
( z − α1 )( z − α 2 )L( z − α p )

Numerator Roots define “zeros” Poles must be inside

Denominator Roots define “poles” unit circle for stability

Poles & zeros must occur in

In this case Complex Conjugate pairs
p–q zeros at z=0 9
to give real-valued TF coefficients
 IIR and FIR Filters

b0 + b1 z −1 + L + bq z − q
H z ( z) =
1 + a1 z −1 + L + a p z − p

FIR: Finite Impulse Response – h[n] has only finite many nonzero values
Filter is FIR: If ai = 0 for i = 1, 2, …, p (It is IIR otherwise)

H z ( z ) = b0 + b1 z −1 + L + bq z − q

bn , n = 0,1,..., q TF coefficients are the

h[n ] =  impulse response values!!
0, otherwise

10
 Example System

Time Domain Z / Freq Domain

Difference ZT (Theory) Transfer

Equation Inspect (Practice) Function

[ ]
Input-Output Form
y (n) − α y (n − 1) = β x(n) Y (z) 1−α z = β X (z)
−1

Recursion Form
y (n) = β x(n) + α y (n − 1)
Y ( z) β
H ( z) = =
X ( z ) 1 − α z −1
11
 Example System (cont.)

Time Domain Z / Freq Domain

x(n) y(n)
β Σ
α z-1
y(n-1)

Block
Inspect Diagram
Inspect

Difference ZT (Theory) Transfer

Equation Inspect (Practice) Function

β
Recursion Form H ( z) =
y (n) = β x(n) + α y (n − 1) 1 − α z −1
12
 Example System (cont.)

Z / Freq Domain

β
H ( z) = Transfer
z = e jΩ 1 − α z −1 Function
z −1 = e − jΩ
On z −1 = e − jΩ Unit Circle
Unit
Circle β Frequency
H (Ω ) = Response
1 − α e − jΩ
Ω ∈ (−π , π ]

13
Example System (cont.)
Euler’s
Plotting Transfer Function Equation

β β
H (Ω) = =
1 − α e − jΩ 1 − [α cos(Ω) − jα sin(Ω)]
Group into
β
= Real & Imag
[1 − α cos(Ω)] + jα sin(Ω)

β
H (Ω) = Standard
[1 − α cos(Ω)] + [α sin(Ω)]
2 2
Eqs for
 α sin(Ω)  Mag. & Angle
∠H (Ω) = − tan −1  
1 − α cos( Ω ) 

14
 Example System (cont.)

Unit
Z/Freq Domain Circle
Im{z}
Pole/Zero
Diagram
Re{z}
Roots

Transfer
Function Zero
Num. = 0
β βz When z = 0
H ( z) = =
1−α z −1
z −α
Pole
If α < 1: Inside UC
Den. = 0
If α > 1: Outside UC
When z = α
15
 Example System (cont.)
Time Domain Z / Freq Domain
β
H ( z) =
Inv. ZT 1 − α z −1
Table
(See Porat)
Transfer
Function

h(n) = βα n
ZT
Unit Circle

Impulse DTFT Frequency

Response Response

β
H (Ω ) =
1 − α e − jΩ
Inv. DTFT
π β
h( n) = ∫ dΩ
−π 1 − α e − jΩ
Ω ∈ (−π , π ]
16
Review of Standard
Sampling Theory
(P-3.1-3.2)

17
 Practical Sampling Set-Up
ADC DAC
x(t)
x[n] = x(nT) Pulse
“Hold” CT LPF
Gen
Sample at ~
t = nT ∞ x r (t )
T = Sampling Interval
~
x p (t ) = ∑ x(nT ) ~p(t − nT )
n = −∞
Fs = 1/T = Sampling Rate

1
Signal Value

-1

-2
0 0 .2 0.4 0 .6 0.8 1
Tim e
2

1
Signal Value

-1

-2
0 0 .2 0.4 0 .6 0.8 1 18
Tim e
 Sampling Analysis (#1)
Goal = Determine Under What Conditions We Have:
Reconstructed CT Signal = Original CT Signal ~
x (t ) = x (t )

Simplify to Develop Theory: Use ~

p (t ) = δ (t )
Why???? 1. Because delta functions are EASY to analyze!!!
2. Because it leads to the best possible case

∞
~
x p (t ) ⇒ x p (t ) = ∑ x(nT )δ (t − nT )
n = −∞

xp(t) is called the “Impulse Sampled” signal

Note: xp(t) shows up during Reconstruction, not Sampling!!

19
 Sampling Analysis (#2)
Now, the impulse sampled signal xp(t) is filtered to give
the reconstructed signal xr(t)
∞ x r (t ) = ???
x p (t ) = ∑ x(nT )δ (t − nT )
n = −∞ b
∞ CT LPF
Xr ( f ) = H( f )X p( f )
= x (t ) ∑δ (t − nT )
n = −∞ H( f )

Need to find Xp( f )…. But First a Trick!!!

{ }
∞ ∞
∑δ (t − nT )
∑
x p (t ) = x (t ) 1 j 2πkFs t
= −∞
n1
X p( f ) = F x ( t ) e
44244
3
periodic ⇒ use FS
T k = −∞
∞ ∞
∑
1 j 2πkt / T
∑ X ( f − kFs )
= x (t ) e 1
T k = −∞
=
∞
T k = −∞
∑ x (t )e
1 j 2πkFs t
=
T k = −∞
20
 Sampling Analysis (#3)
ADC DAC
x(t)
x[n] = x(nT) Impulse
“Hold” CT LPF
Gen
Sample at
t = nT x p (t ) x r (t )

X( f )

–B B
f
Xp( f )

–2Fs –Fs Fs 2Fs

f
H( f )

f
Xr( f )

Xr( f ) = X( f ) f
If Fs ≥ 2B 21
 Sampling Analysis (#4)
What this says: Samples of a bandlimited signal completely define
it as long as they are taken at Fs ≥ 2B

Impact: To extract the info from a bandlimited signal we only need to

operate on its (properly taken) samples

# Use computer to process signals

x(t)
x[n] = x(nT) Extracted
“Hold” Computer
Sample at Information
t = nT

22
 Sampling Analysis (#5)
FT of Impulse Sampled Signal gives view of Original FT
&
FT of Impulse Sampled Signal = DTFT of Samples

# DTFT gives view of FT of original signal

Let’s see this

 ∞  ∞
F {x p (t )} = F  ∑ x ( nT )δ (t − nT )  = ∑ x ( nT ) F {δ (t − nT )}
n = −∞  n = −∞ 14 4244 3
e − jnωT
∞ ∞
= ∑ x ( nT ) e − jnωT
= ∑ x[ n ]e − jnθ
θ = ωT
n = −∞ n = −∞

F f
X p (θ / T ) = X (θ ) 23
 Sampling Analysis (#6)

X( f ) θ = ωT
–B B f
 2π 
DTFT
=  f
–2Fs –Fs –Fs/2 Fs/2 Fs 2Fs
f  Fs 
θ DT Freq
–4π –2π –π π 2π 4π (rad/sample)

θ/π Normalized DT Freq

–4 –2 –1 1 2 4

Read notes on web on “Concept of Digital Frequency”

24
 Introduction to
EE521 Case Study

25
 EE521 Case Study: Emitter Location
Processing Tasks
• Intercept RF Signal @ Rx’s
Radio Freq Transmitter (Tx)
• Sample Signal Suitably for Processing • Communications
• Detect Presence of Emitter’s Signal • Radar
• Estimate Characteristics of Signal
• Use Est’d Char’s to Classify Emitter
• Share Data Between Rx’s
• Cross-Correlate Signals to Locate Tx

Data
Link
Receiver #3
Data Link (Rx3)
Receiver #1
(Rx1) Receiver #2
(Rx2) 26
 Overall Processing Set-Up
Rx1
Antenna Estimate
Detect Classify
Signal
Analog ← → Digital Signal Emitter
Parameters

RF Sampling Digital Compress

Front-End Sub-System “Front-End” & Data Link

RF Data Link
Rx2
Antenna Data Link &
Analog ← → Digital Decompress

RF Sampling Digital Cross

Front-End Sub-System “Front-End” Correlation

Estimate
Detect Classify
Signal
Signal Emitter
Parameters 27
 Processing Tasks Overview (#1)
RF Front-End (Analog Circuitry): “RF” = “Radio Frequency”
• Selects reception band; may need to scan bands
• Amplifies signal to level suitable for ADC, etc.
• Frequency-shifts signal spectrum to range suitable for ADC
! Technology trend is toward requiring less shifting
• Unfortunately: noise introduced by analog electronics (Random Signals)
Topic We’ll Cover

FT of Signal
Before RF-FE

1 GHz 3 GHz 4 GHz

Antenna

FT of Signal
RF After RF-FE
Front-End
200 MHz 28
 Processing Tasks Overview (#2)
Sampling Sub-System (Mixed Analog/Digital Circuitry):
• Converts analog CT signal into digital DT signal (Bandpass Sampling)
• Converts real-valued RF signal into complex LP signal (Equiv. LP Signals)
Digital Front-End (Digital Processing):
Topics
• Convert to complex LP signal (Equiv. LP Signals) We’ll
• Equalizes response of analog front-end (DFT-Based Filtering) Cover
! Magnitude and Phase
• Bank of digital filters splits signal into sub-bands
for further processing (Filter Banks)
• Reduce sampling rate after filtering to sub-bands (Multi-Rate Processing)
Signals are
Oversampled

DTFT of Signal
–π π θ
Before Digital FE

Digital
–π π θ
“Front-End”

–π π θ rad/sample
29
–π π θ
 Processing Tasks Overview (#3)
Detect Presence of Signal (Digital Processing):
• Has signal been intercepted or only noise in subband (DFT-Based Proc.)
Estimate Parameters of Signal (Digital Processing) (see also EE522):
• Estimate frequency of signal (DFT-Based Processing)
Classify/Model Signal (Digital Processing):
• What type of signal is it? (Spectral Analysis of Random Signals)
PSD Model
 b0 + b1 z −1 + b2 z −2 + L + b p z − p 
Px (θ ) = σ 2  −1 −2 −q

–π π θ  1 + a1 z + a 2 z + L + a q z  jθ
z =e
Use PSD Model parameters {bi} and {ai} to classify signal type

Compress/De-Compress Signal (Digital Processing) (see also EE523):

• Exploit signal structure to allow efficient transfer
(DFT-Based Proc./ Filter Banks / Spectral Analysis)
Cross-Correlate Signals (Digital Processing):
• Compute relative delay and Doppler (DFT-Based Proc.& Multi-Rate Proc.)
30