Digital Carrier Systems

EE 442 – Spring Semester

Lecture 12

ON-Off Keying
1 0 1 1 0 1

1 0 0 1 0 1

(4 states)

1
 Digital Carrier Systems

In the last lecture we studied “baseband” digital signals; that is, the
modulating signal m(t) have not been frequency shifted.

However, for wireless and satellite communications we must use higher

frequencies to transmit and receive communication signals.

Now we require a modulator and a demodulator – together they form

a “modem.”

There are two basic forms of carrier modulation – they are (1) amplitude
modulation and (2) angle modulation (phase and frequency modulation).
We have already studied both under the heading of analog modulation.

2
 Example: Amplitude Shift Keying (ASK)

This is binary amplitude shift keying (BASK).

m(t)

 ASK = m(t ) cos (C t )

3
 Example of Multilevel ASK with 2-Bit Coding

This is multilevel amplitude shift keying.

Bit 1
Bit 2

Symbols 00, 01, 10 & 11 translate into four amplitude levels.

http://www.tmatlantic.com/encyclopedia/index.php?ELEMENT_ID=10420

4
 Band Limiting Softens the Edges of ASK Waveforms

This is the more realistic case for ASK communication systems.

In fact, all waveforms are softened by bandwidth limitations
of transceiver equipment and the channel itself.

m(t)
time

Notice the similarity between ASK and analog AM because the amplitude
of the modulated signal is proportional to m(t).

http://www.slideshare.net/Zeolite27/dc-ppt-final
5
 Next: Phase Shift Keying (PSK)

Angle modulation gives rise to both phase modulation and frequency modulation.
Starting with phase modulation; this is generally known as “phase shift keying.”

Example: m(kTb) = +1

m(kTb) = -1

http://electronicdesign.com/communications/understanding-modern-digital-modulation-techniques

6
 Constellation Diagram For PSK

AC  cos(C t ) for m( kTb ) = +1

AC  m(t ) cos(C t ) =
AC  cos(C t +  ) for m( kTb ) = −1
Q

PSK
A special case: on-off keying (OOK)
Q
I
OOK
0
I

We can also express as I and Q components.

7
 Expressing PSK in I and Q Components

For PSK we can write,

 PSK = AC cos (C t +  k ) for kTb  t  kTb + Tb

 PSK = AC cos( k ) cos (C t ) − AC sin( k )sin (C t )

Therefore,
 PSK = I k cos (C t ) + Qk sin(C t ) for kTb  t  kTb + Tb

This is in polar form (I and Q)

For binary PSK we have  k = 0 or  radians – there is no Q component.

This is 2-QAM, but we don’t use this terminology for binary PSK.

Note: Quadrature amplitude modulation (QAM) is a mixture of both

amplitude modulation and phase modulation.

8
 Binary PSK (BPSK) Transmitter and Receiver
BPSK Modulator:
Carrier Balanced
Amplifier
cos(ct) Modulator BPF
PSK

NRZ Data LPF

input

BPSK Demodulator:
LPF S&H Comparator
PSK
+ d(t)

Sample at Binary data

center of output
cos(ct) symbol

r (t ) B cos[2C t +  (t )] + B cos[ (t )]
9
 Binary PSK (BPSK) Waveforms at the Receiver

Without noise With noise

After Lawrence Burns, “Digital Modulation and Demodulation,” Chapter 4

in RF and Microwave Circuit Design for Wireless Communications, edited
by Lawrence E. Larson, Artech House Publishers, 1996. Pages 99 to 233.
Lawrence Burns was an engineer at 3COM.

10
 BPSK Waveforms and Noise

Sampled data points:

Constellation Diagram

11
 Frequency Shift Keying (FSK)

Frequency Shift Keying (FSK):

Bit 1

Bit 0

Continuous-Phase Frequency Shift Keying (CPFSK):

Bit 1

Bit 0
1 0 1 0

aka MSK

https://www.cdt21.com/resources/Modulation/modulation_FSK.asp
12
 Minimum Shift Keying (MSK)
Minimum shift keying (MSK) is a form frequency modulation based on a
system called continuous-phase frequency-shift keying (CPFSK).
MSK advantages are (1) better spectral efficiency as compared to other
modes, and (2) it allows power amplifiers to operate in saturation
enabling higher levels of efficiency.

https://www.electronics-notes.com/articles/radio/modulation/what-is-msk-minimum-shift-keying.php
13
 Minimum Shift Keying (MSK)

For Bit 0 (frequency f 0 ) we write,

2Eb
s0 (t ) =
Tb
(
cos 2 f 0 t ) over interval 0  t  Tb

For Bit 1 (frequency f 1 ) we write,

2Eb
s1 (t ) =
Tb
(
cos 2 f 1 t ) over interval 0  t  Tb

Tb

For orthogonality we require  s (t )s (t )dt = 0

0
1 2

1
Define the frequency separation: f = f 1 − f 2 =
2Tb
is the minimum separation to ensure orthogonality to hold.
f +f
The carrier frequency fC is defined to be fC = 0 1
2

14
 MSK Modulation and Demodulation
MSK Modulator:
NRZ Data Voltage
input Controlled Amplifier
Oscillator BPF
t FSK
Vcontrol RF
VCO Output

MSK Demodulator:
BPF Amplifier LPF Comparator
FSK
+ m(t)

Frequency Binary data

Discriminator output

15
 Gaussian Minimum Shift Keying (GMSK)

Gaussian Minimum Shift Keying (GMSK) modulation is a modified

version of the Minimum Shift Keying (MSK) modulation where the
phase is further filtered through a Gaussian filter to smooth the
transitions from one point to the next in the constellation of states.

A Gaussian filter’s impulse response is a Gaussian function (or an

approximation to it, because a true Gaussian response is physically
unrealizable). Gaussian filters have no overshoot in responding to a
step function input and minimize the rise and fall times of pulses.

GMSK
waveform

http://home.scarlet.be/~pc030062/extra_info/MSK%20-%20GMSK.pdf
16
 Gaussian Minimum Shift Keying (GMSK)
Advantages:

(1) Constant amplitude (better noise resistance)

(2) Spectrally efficient (spectrum falls as 4th power of frequency)
(3) Good Bit Error Rate performance
(4) Self-synchronizing capability

Applications:
(1) Automatic Identification System in maritime navigation
(2) Bluetooth headsets
(3) Standard GSM cellular (usage is > 40% of cell phones in world)
Uses BTb = 0.3 for Gaussian filter giving 99% of
transmission power in bandwidth of 250 kHz. The GSM
bandwidth per channel is 200 kHz and GSM transmits
at 270 kb/sec (limited by ISI problems). B = 81.3 kHz
and Tb = 3.7 microseconds.

Tradeoff: Smaller BTb product gives faster spectrum falloff at

expense of longer pulse duration leading to increased ISI.
17
Comparing Spectrums of MSK with GMSK (BTb = 0.5 and 0.3)

MSK

Spectral efficiency = 1.35 bps/Hz

https://en.wikipedia.org/wiki/Minimum-shift_keying#/media/File:GMSK_PSD.png
18
 Generating GMSK Signals

GMSK modulator using VCO

GMSK modulator using an I-Q modulator

https://www.electronics-notes.com/articles/radio/modulation/what-is-gmsk-gaussian-minimum-shift-keying.php
19
Second Generation (2G) Mobile Phones
❑ Second Generation (2G) introduced in early 1990s to replace 1G
❑ Digital (rather than analog) transmission of voice
❑ Designed for circuit-switched networks (voice centric standard)
❑ Uses 900 MHz and 1800 MHz frequency bands
❑ Dominant 2G phone standard: GSM (Global System for Mobile)
❑ Introduced SMS (aka “text messaging”)
❑ GSM uses SIM card containing user ID (Subscriber Identity Module)
❑ Began in Europe; rapidly expanded around the World

GSM is still
widest used
cell phone
standard
Worldwide!
Nokia 2G
http://sharingmythoughts-
ben.blogspot.com/2010/09/histor Mobile
y-of-mobile-phone.html Phones
http://en.wikipedia.org/wiki/Global_System_for_Mobile_Communications
20
 Comparing PSDs For Binary ASK, PSK and FSK

ASK

Power spectral density [watts/Hz)

PSK

FSK

Figure 7.32 in Lathi & Ding, Modern Digital and Analog Communication Systems. 4th ed., 2009.
21
 Multilevel Frequency Shift Keying (FSK)

This animation shows frequency Symbol Binary code Frequency

shift keying of the sinusoidal “0” 00 4 kHz
carrier signal. A two-digit code “1” 01 3 kHz
modulates the carrier signal “2” 10 2 kHz
frequency into four frequencies “3” 11 1 kHz

22
 Quadrature Phase Shift Keying (QPSK)

Sometimes this is known as quadrature-phase PSK, 4-PSK, or 4-QAM.

QPSK uses a circle of four points on the constellation diagram, equi-spaced
around a circle. With four states, QPSK can encode two bits per symbol,
Q

I = -1; Q = +1 I = +1; Q = +1

QPSK = I  cos (C t ) + Q  sin (C t ) I

I = -1; Q = -1 I = +1; Q = -1

23
 Digital I/Q Modulation
Anticipating our coverage of digital communication systems

i(t)

q(t)

24
 Simple QPSK Modulator

QPSK modulator using delay lines to set phase delay:

Delay lines (depend upon fC)

+45

RF Input +135 RF Output

-135

-45

Switch Decoder and Driver

25
 Widely-Used QPSK Modulator

QPSK Modulator

LPF
I
NRZ Data Amplifier
input BPF
Serial-to PSK
-Parallel
Parser
cos(C t ) 
LPF

t
Q

sin(C t )

26
 Basic Building Block: Quadrature Modulator

I and Q can be
either analog or cos(C t ) 
digital signals  (t )
Q

sin(C t )

 (t ) = I 2 + Q 2  cos(C t +  (t ))
 Q(t ) 
where  (t ) = tan −1  
 I ( t ) 

27
 Data Demultiplexer (Serial to Parallel) For QPSK

Demodulator uses three D-type flip-flops and is driven by both clock

and one-half clock rates.

28
 QPSK Demodulator

PSK(t)

C/R = clock/carrier recovery

STR = symbol timing recovery

29
 M-ary Signaling With Quadrature Amplitude Modulation (QAM)

APSK definition
Definition: Amplitude and Phase-Shift Keying, APSK, is a digital
modulation scheme that uses both the amplitude and the phase changes
of on the carrier signal to provide the data transport mechanism for the
information. Also called QAM.

16-ary QAM
Quadrature Amplitude
Modulation, QAM is a
form of modulation that
is a combination of
phase modulation and
amplitude modulation.
The QAM scheme
represents bits as
points in a quadrant
grid know as a
constellation map.

30
 Number-Bases in M-ary Constellations

(aka 4-QAM)

Variants of QAM are also used for many wireless and cellular technology
applications. In addition, 64-QAM and 256-QAM are commonly used in digital
cable television and cable modem applications. In the US, 64-QAM and 256-
QAM are the mandated modulation schemes for digital cable as standardized by
the SCTE in the standard ANSI/SCTE 07 2000.

SCTE = Society of Cable Telecommunications Engineers

31
 Bits/Symbol and Symbol Rates

Bits per
Modulation Symbol Rate
Symbol
BPSK 1 1  bit rate
QPSK 2 1/2 bit rate
8-PSK 3 1/3 bit rate
16-QAM 4 1/4 bit rate
32-QAM 5 1/5 bit rate
64-QAM 6 1/6 bit rate

32
Greater Number of States Leads to Greater Demand Upon
Communication System

http://farhek.com/jd/i1t1154/up-to/7i45u1/
33
Bit Error Rate versus Energy/Noise Ratio

BER = Bit Error Rate

Eb / N0 (dB)
energy per bit-to-noise power ratio

34
 Signal-to-Noise Ratio vs. Energy/Bit-to-Noise Ratio

In analog and digital communications, signal-to-noise ratio, usually written S/N

or SNR, is a measure of signal strength relative to background noise strength.
The ratio is usually expressed in decibels (dB) and equals 10log10[S/N].

Another metric that is often more useful in digital systems is the energy per
bit-to-noise power ratio, denoted by Eb/N0.
Define: Rb = bit rate (in bits per second)
S = total signal power (watts)
Eb = energy per bit (in joules/bit)
N = total noise power (over entire bandwidth B in Hz)
N0 = noise spectral density (N = N0B where B = bandwidth)
Then,
S Eb S Rb Eb
= Eb and = and SNR =
Rb N Rb  N N0B

Increasing the data rate Rb increases the SNR. However, in general it also
increases the noise in the denominator, which lowers the SNR.

35
 What Modulation Schemes Does Wi-Fi Use?

WiFi systems use two primary radio transmission techniques.

802.11b (≤ 11 Mbps) − The 802.11b radio link uses a direct sequence spread
spectrum technique (DSSS) called complementary coded keying (CCK). The
bit stream is processed and then modulated using Quadrature Phase Shift
Keying (QPSK).

802.11a and 802.11g (≤ 54 Mbps) − The 802.11a and g systems use 64-
channel orthogonal frequency division multiplexing (OFDM). The
transmitter encodes the bit streams onto 64 subcarriers using Binary Phase
Shift Keying (BPSK), Quadrature Phase Shift Keying (QPSK), or one of two
levels of Quadrature Amplitude Modulation (16-QAM, or 64-QAM).

36
Circuit Switched Networks vs. Packet-Switched Network

https://www.photonics.com/Articles/Toward_Optical_Packet_Switching/a24582

37
 Circuit-Switched Network
PSTN = public switched telephone network
Once a connection is
Caller established, this
connection is
Central Office maintained until call
Telephone is terminated.
Telephone Switch
Many paths are Switch
possible, but only Full Telephone
one is selected per Duplex
Switch
call. Telephone
Switch
Trunks
Telephone (links between
Switch Exchanges) Telephone
Subscriber lines
(or local loops) Switch
Central Office Telephone
Switch Central Office

= Dedicated connection (point-to-point)

38
 Packet Switched Network
Packet route
Message broken
Many paths possible
into packets and
for a single message as = Packet
each addressed
packets are routed to
the destination. (Data Packet or “Datagram”)

Router
or Switch
Sender
(source) Internet

Packets are routed Packets

according to the best sequentially
path available at the reassembled
time. to reveal
Large array of routers and data links. message
Receiver
(destination)

39
 Network Organization
In 1962, Paul Baran (RAND Corp.) envisioned a network of unmanned nodes using
intelligent switches to route data node to node to their final destinations. Baran
called this "hot-potato routing" or distributed communications. This was
implemented in ARPANET which became the Internet.
A network
of routers

Centralized Network Decentralized Network Distributed Network

A highly vulnerable
network (e.g., PSTN) (e.g., Internet)
Concept of hardened networks to deal with disasters. 40
Internet is a “Network of Networks”
A “representative” section of the Internet: Packet Switch Modem
or Router Cellular
(ISP = Internet Service Provider) Wireless
Base Station
Router
Key: VoIP Phone
Example: Cell Phone
Verizon Host
Mobile
National Laptop Server

& Global ISPs

Corporate or
Mobile University
Cellular Wireless Example:
Network
Network Sprint

Home
Network

Local or
Home Regional
Network ISP

Example:
Example: Keysight Technologies
sonic.net

After Kurose & Ross, Computer Networking, 5th Edition, Addison-Wesley, New York, 2010. 41
 Comparing Packet Switching to Circuit Switching
Feature Circuit Switching Packet Switching
Dedicated Path No
Yes

Route is established on a per

Path dedicated for packet basis of the
Path Formation one conversation conversation using datagram
(or per conversation with
virtual circuit)
Packet transmission delay
Delay Call setup delay (call setup delay for virtual
circuit)
Bandwidth Type Fixed Bandwidth Dynamic bandwidth
Increases packet delay (can
Stops call block call establishment and
Overload Effects
establishment increase packet delay with
virtual circuit)

42
 Packet-Switched Network Operation

• Adaptive routing – routers chose the best path by examining

traffic loading along available paths. Routers create a
“routing table” for the packet travel.
• All users share the same network resources.
• Packet-switching is more efficient than circuit-switching in
networks when data is bursty (i.e., variable delays
interspersed with periods of data transmission). More
“efficient” means a better utilization of the network
resources.

This is an
example of
“bursty” data

43
 TCP/IP Protocol Architecture Model

Layer Pictorial View of Protocol Data Unit Entity Protocol

Data or SMTP
Application Data Message HTTP, DNS

Number of segments  1 Transport TCP

Transport Header
Data Segments
UDP

Internet Network Transport Packets or

Header Header Data IP
or Network Datagrams

Frame Network Transport

Data
Frame Ethernet
Network Header Header Header Trailer
Frames Modem
Access
FDDI
Bits transmitted over channel medium

44
 An Internet Packet and its Headers
Internet Packet
IP header TCP/UDP header Application Data

• In IPv4, each packet is restricted to 1,500 bytes of data (i.e., payload)

• Each packet consists of the application data and headers
• The headers contain control and routing information such as:
– Source IP address and destination IP address
– Packet numbering for reconstruction at destination
• Every computer on the Internet has the TCP/IP program. The client/server model
is used on the Internet.
• TCP (Transmission Control Protocol) puts the data or message into packets at the
source and reassembles the data or message at the destination
• IP (Internet Protocol) does the packet addressing for the routing over the Internet

The rules that govern communication – any form – are called “protocols.”
45
 TCP versus UDP Transmission

TCP and UDP Analogies:

TCP Post Office
Registered Verifies delivery
Letter
TCP is “reliable” because
it has flow & congestion
control, retransmission, &
uses acknowledgements.

UDP does not use these UDP

because it is focused only
upon sending packets.

46
 IPv6

How do you say 340,282,366,920,938,463,463,374,607,431,768,211,456?

It is 340 undecillion, 282 decillion, 366 nonillion, 920 octillion, 938

septillion, 463 sextillion, 463 quadrillion, 607 trillion, 431 billion,
768 million, 211 thousand, 456.
47
 Internet Attributes
▪ Uses computer-based technologies
▪ Highly interconnected but distributed in practice
▪ Networks interconnected thru gateways and access
points (i.e., routers, link-layer switches, etc.)
▪ Each network stands on its own (TCP/IP protocols do not
dictate internal changes to networks)
▪ TCP/IP is independent of the data type or the transport
medium
▪ TCP/IP protocols are not proprietary (all can freely use it)
▪ Effectively little regulation applies to Internet (in fact,
users of the Internet generally oppose regulation)
▪ There is NO direct global control of the Internet

48
Questions

49
Bandwidth Efficiency (aka Spectral Efficiency)

Given: Eb = energy per bit (joules or ergs)

Rb = bit rate (bits/second)
B = bandwidth of baseband signal (Hz)
N0 = noise spectral density (watts/Hz)
N = noise power = N0B (watts)

Therefore, EbRb = total signal power

We define the Bandwidth Use Efficiency as

Rb  bits/second  Example:
B  Hz  GSM Digital Cellular
Data rate = 270 kb/s
In general, B = 200 kHz, thus
Rb  ER  Bandwidth efficiency =
= log 2  1 + b b  1.35 bits/sec/Hz
B  NB 

50
http://wiki.ucalgary.ca/page/Courses/Computer_Science/CPSC_441.W2014/Ch
apter_1:_Computer_Networks_and_The_Internet.html
51