Chapter 3

Bandpass Transmission Techniques

for Wireless Communication

ECE414 Wireless Communications, University of Waterloo, Winter 2012

Outline
Introduction to Digital Communications
Signal (Vector) Space Representations
Digital Modulation Schemes (M-ASK, M-PSK, M-FSK)
Performance Measures for Modulation Schemes
- Bandwidth (spectral) efficiency
- Power efficiency

- Temporal characteristics (e.g., dynamic power range, peak/average ratio)

Power Spectral Density of Digital Modulation Schemes
Error Rate Performance of Digital Modulation Schemes

Comparison of Digital Modulation Schemes in terms of Spectral

Efficiency and Power Efficiency
Temporally Efficient Digital Modulation Schemes
ECE414 Wireless Communications, University of Waterloo, Winter 2012

Block Diagram for a Digital Communication System

Original
message signal
(analog)

A/D

Source
Encoder

Channel
Encoder

Modulator

Channel

Recovered
message
signal (analog)

D/A

Source
Decoder

Channel
Decoder

Demodulator

Analog-to-Digital (A/D) Conversion: Analog (i.e., continuous-time

continuous-amplitude) message signal is converted into a discrete-time discreteamplitude digital signals by time-sampling and amplitude-quantization. The
resulting signals are then mapped to binary sequences.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

Block Diagram for a Digital Communication System (contd)

Source Encoding: Removes the redundant information embedded in the
message signal, therefore represents the message with as few binary digits as
possible, i.e., data compression
Channel Encoding: Introduces redundancy in a controlled manner which can
be used at the receiver to overcome the effects of noise, interference and fading.
Provides noise immunity to transmitted information.
Source coding and channel coding will not be studied in this course

Modulation: Converts (maps) codewords to high-frequency analog waveforms.

A certain parameter of the carrier signal (i.e., modulated signal) is varied in
accordance with message signal (i.e. modulating signal) e.g. amplitude shift keying
(ASK), phase shift keying (PSK), frequency shift keying (FSK)
Receiver Blocks: Perform the inverse of the transmitter operations in order to
recover the original analog message (continuous-time continuous-amplitude) signal.
In a practical digital communication receiver, there are also additional sub-blocks
such as channel estimation, synchronization (frame/frequency/phase),
authentications, crypto, multiplexing, etc.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

Why is Modulation Required?

Modulatio shifts the baseband signal to a higher frequency band, centered at the socalled carrier frequency.
To achieve easy radiation: Dimensions of the transmit/receive antennas are
limited by the corresponding wavelength. The frequency conversion allows the
use of practical antenna lengths.
To accommodate for simultaneous transmission of several baseband
signals: Simultaneous transmission of different baseband signals which are
possibly overlapping can be facilitated by assigning slightly different frequency
carriers for each one.
Large bandwidths require high carrier frequencies: Practical requirements
in front-end filter design dictates the bandwidth-to-frequency carrier ratio (i.e.,
fractional bandwidth) be kept within a certain range.

B
0.01 0.1
fc

B
: Fractional bandwidth
fc
ECE414 Wireless Communications, University of Waterloo, Winter 2012

Why is Modulation Required? (contd)

High-rate transmission requires larger bandwidths (therefore, higher
carrier frequencies): According to Shannon Theorem, channel capacity is
defined as the maximum achievable information rate that can be transmitted
over the channel. For the additive white Gaussian noise (AWGN) channel,
C B log 2 1 SNR

Channel capacity

Bandwidth

Signal-to-noise ratio

To (possibly) expand the bandwidth of the transmitted signal for

better transmission quality: When the bandwidth increases, the required
SNR (for fixed noise level, corresponding signal power) to achieve a specific
transmission rate decreases
C
B

C B log 2 1 SNR SNR 2 1

ECE414 Wireless Communications, University of Waterloo, Winter 2012

Signal-Space Representations
Consider a modulation format where the transmitted signal waveforms belong to
M
the modulation set sm t m
1 0 t Ts .

Each of the waveform can be represented as a point (vector) in an N-dimensional

signal space (sometimes called as vector space) defined by the orthonormal basis
N
functions n t n1 N M
T
*
0 s i t j t dt i j

Ts

sm,n sm t
0

n*

t dt st sm,nn t
n 1

sm t s m sm,1, sm,2 ,..., sm, N

The Gram-Schmidt procedure (See Appendix A of the textbook) provides a
systematic approach to construct the set of orthonormal functions, which span the
signal space.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

Signal-Space Representations (contd)

sm t sm sm,1, sm,2 ,..., sm, N
Energy

Es

Ts

2
sm
0

t dt sm2 ,n sm

sk t sk sk ,1, sk ,2 ,..., sk , N
sl t sl sl ,1, sl ,2 ,..., sl , N
Correlation

Euclidean
Distance

n1

k , l m 1,2,...M

1
sk ( t ), sl ( t ) =
Ts
d ( sk ( t ), sl ( t )) =
2

Ts

s (t )s (t )
k

dt = sk , sl = s k sl = sk,n s*l,n
n=1

Ts

sk ( t ) - sl ( t ) dt = d ( s k , sl ) = sk,n - sl,n
2

ECE414 Wireless Communications, University of Waterloo, Winter 2012

n=1

M-ary Amplitude Shift Keying (M-ASK)

Am 2m 1 M , m 1,2...M

sm t Am cos2f c t

0 t Ts

Baseband (Equivalent Lowpass) Representation

sm t Am cos2f c t Re Am e j 2fct sm,lp t Am

Basis Function(s) (Obtained through Gram-Schmidt procedure)

2 Ts cos2f ct , 0 t Ts

0,

otherwise

Signal-Space (Vector Space) Representation (Obtained through the use of

basis functions)

sm t sm Am Ts 2

1-dimensional

Ts

Signal Energy Esm sm2 t dt Am2 Ts 2

0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

M-ASK (contd)
Examples of M-ASK
Signal Constellations

M=4
Bandpass Modulation Signal
11 10 00 01

Equivalent Lowpass Signal

ECE414 Wireless Communications, University of Waterloo, Winter 2012

10

M-ary Phase Shift Keying (M-PSK)

sm t A cos2f c t m

m 2 m 1 M , m 1,2...M

0 t Ts

Baseband (Equivalent Lowpass) Representation

sm t A cos2f c t m Re Ae jm e j 2fct sm,lp t Ae jm

Basis Functions
2 Ts cos2f ct , 0 t Ts
2 Ts sin 2f ct , 0 t Ts
1 t
2 t
otherwise
otherwise
0,
0,

Signal-Space Representation

sm t s m A Ts 2 cos m , A Ts 2 sin m

2-dimensional

Ts

Signal Energy Esm Es sm2 t dt A2Ts 2

0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

11

Example: Binary Phase Shift Keying (BPSK)

1 s1 t A cos2f ct
0 s2 t A cos2f c t
Signal-Space Representation

Bandpass Modulation Signal

2 t

Es

Es

1 t

-A
Equivalent Lowpass Signal

s1 Es ,0

s 2 Es ,0

A
t
-A

ECE414 Wireless Communications, University of Waterloo, Winter 2012

12

Example: Quadrature Phase Shift Keying (QPSK)

s1 Es ,0

00 s1 t A cos2f ct

s 2 0, Es

01 s2 t A cos2f ct 2

10 s3 t A cos2f ct s 3 Es ,0

s 4 0, Es

11 s4 t A cos2f ct 3 2

2 t

Signal-Space
Representation

Es
Es

Es

1 t

Es

ECE414 Wireless Communications, University of Waterloo, Winter 2012

13

Quadrature Amplitude Modulation (QAM)

sm t Am,r cos2f ct Am,i sin2f ct m 1,2,...M

0 t Ts

Am,r, Am,i: Information-bearing signal amplitudes of the quadrature carriers

Alternatively, QAM can be considered as a combination of ASK and PSK.
sm t Am cos2f ct m where

Am Am2 ,r Am2 ,i

m 1,2,...M

0 t Ts

m arctgAm,i Am,r

Examples of QAM
Signal Constellations

ECE414 Wireless Communications, University of Waterloo, Winter 2012

14

QAM (contd)
Baseband (Equivalent Lowpass) Representation

sm t Am,r cos2f ct Am,i sin 2f ct Re Am,r jAm,i e j 2fct

sm,lp t Am,r jAm,i Am e jarctgm

Basis Functions
2 Ts cos2f ct , 0 t Ts
1 t
otherwise
0,

2 Ts sin 2f ct , 0 t Ts
2 t
otherwise
0,

Signal-Space Representation

sm t s m Am,r Ts 2, Am,i Ts 2

2-dimensional

Signal Energy
Ts

E sm sm2 t dt Am2 ,r Am2 ,i Ts 2

0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

15

M-ary Frequency Shift Keying (M-FSK)

sm t A cos2 f c f m t

f m mf , m 1,2...M

0 t Ts

Baseband (Equivalent Lowpass) Representation

sm t A cos2 f c f m t Re Ae j 2fmt e j 2fct sm,lp t Ae j 2fmt

Cross Correlation

k ,l

1 Ts j 2 k l ft
sinT k l f jT k l f
e
dt
e
Ts 0
T k l f

k ,l

sin2T k l f
2T k l f

For f 1 2T and k l

k ,l 0

Therefore, the minimum frequency separation between adjacent signals for

orthogonality of the M signals is f 1 2T
ECE414 Wireless Communications, University of Waterloo, Winter 2012

16

M-FSK (contd)
Assuming frequency separation f 1 2T, the signal-space representation
for the M-FSK signals are given as N-dimensional vectors, where N=M.
2 Ts cos2 f c f m t , 0 t Ts

m t

0,

otherwise

s1 t s1 Es

0 0 ...... 0

s2 t s2 0

Es

0 ...... 0

.
.
.

sM t s M 0 0 0 ......

Es

Ts

where Esm Es sm2 t dt A2Ts 2

0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

17

Performance Measures for Modulation Schemes

Bandwidth (spectral) efficiency: How much bandwidth is needed for a
given data rate?
s : Bandwidth efficiency
Rs log 2 M Ts
Bits/sec/H z
s

Rs : Data rate W : Bandwidth

W
W
The bandwidth depends on the modulation scheme and pulse shaping. Power
spectral density (PSD) is typically used to determine the bandwidth of the
transmitted signal. There are various definitions for bandwidth:

Main lobe (null-to-null) bandwidth: The width of the main spectral lobe.
Fractional power-containment bandwidth: The frequency interval that
contains (1-) of the total signal power, e.g. 99.9% of the total power.
Bounded PSD bandwidth: The frequency interval where the PSD stays
above a prescribed certain threshold, e.g. sidelobes peaks 40 dB below its
maximum value
Roughly speaking, bandwidth of the modulation scheme is proportional to
the dimension number.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

18

Performance Measures for Modulation Schemes (contd)

Power efficiency: How much power is needed for reliable transmission with a
specified fidelity?
The fidelity for a digital communication system is usually measured in terms of
symbol- or bit-error probability. For a given SNR, we aim to achieve a low error
probability (how low? it depends on the application).
Symbol error probability (SEP) is in general easier to evaluate. Bit error
probability (BEP) depends on the mapping of source bits onto modulation signals.
A bound on BEP is given as

Pe
Pb e Pe
log2 M

Two common mapping forms are natural mapping and Gray mapping.
In Gray mapping, the neighbour points differ in only one digit. It should be
noted that Gray mapping is not possible for every signal constellation.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

19

Performance Measures for Modulation Schemes (contd)

Temporal efficiency: How wide are the time variations of the transmitted signal?
Temporal efficiency=Peak power/Average power

The choice of amplifier depends on the temporal characteristics of the signal.

Other considerations:
Hardware/software implementation complexity & cost of implementation

Sensitivity to interference
Robustness to impairments encountered in a wireless channel

In most practical scenarios, these performance measures conflict with each

other. The communication system designer should be able to find the best tradeoff for a given application under specific constraints.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

20

Comparison of Spectral Efficiency of Modulation Schemes

M-PSK and QAM

M-FSK

log 2 M
bits/sec
Data rate
T

Data rate

2
BW null to null Hz
T

BW roughly

Data rate 1
hs =
= log 2 M [ bits/sec/Hz]
BW
2

hs =

log 2 M
bits/sec
T
M
Hz
2T

Data rate 2 log2 M

=
[ bits/sec/Hz]
BW
M

M: Modulation order, Constellation size

ECE414 Wireless Communications, University of Waterloo, Winter 2012

21

Power Spectral Density (PSD)

In practical, pulse shaping should be considered for a precise bandwidth
measurement and considered in the spectral efficiency calculations.
Power spectral density (PSD) describes the distribution of signal power in
the frequency domain. If the baseband equivalent of the transmitted signal
sequence is given as

g t ak pt kTs

a k : Baseband modulation symbol

Ts : Signal interval

pt : Pulse shape

then the PSD of g(t) is given as

g f

1
2
P f a f
Ts

where

See Ch.4 of Digital Communications

by Proakis for the proof

P f F pt

a f Ra n e j 2fnTs
n

1
*
Ra ( n) = E ak ak+n

ECE414 Wireless Communications, University of Waterloo, Winter 2012

22

Example: PSD of BPSK with Rectangle Pulse Shaping

Baseband equivalent of BPSK sequence

g t ak pt kTs

ak A Independent data symbols are assumed

Autocorrelation of data sequence

Ra n E

ak ak*n

E
a
,
n 0 A2 , n 0
k

*
0, n 0

Eak E ak n , n 0

a f F Ra n Ra n e j 2fnTs A2
n

Pulse shaping
t - Ts / 2 FT
- j 2 p f (Ts
p (t ) =
P ( f ) = Tssinc ( fTs ) e
Ts

p(t)

T/2 T

P ( f ) = Tssinc ( fTs )

ECE414 Wireless Communications, University of Waterloo, Winter 2012

23

2)

Example: PSD of BPSK with Rectangle Pulse Shaping (contd)

PSD of baseband BPSK sequence
Fg ( f ) =

P( f )
Ts

Fa ( f ) = A 2Tssinc2 ( fTs )

ECE414 Wireless Communications, University of Waterloo, Winter 2012

24

Example: PSD of BPSK with Rectangle Pulse Shaping (contd)

Bandpass BPSK sequence and its Fourier transform (spectral density)

1
FT
j 2fct

G f fc G* f fc See

S
f

s t Re g t e

Tutorial 1

PSD of bandpass BPSK sequence

1
See Ch.4 of Digital Communications
g f f c *g f f c
4
by Proakis for the proof
1
1
A2Ts sinc2 f f c Ts A2Ts sinc2 f f c Ts
4
4

s f

Null-to-null bandwidth

ECE414 Wireless Communications, University of Waterloo, Winter 2012

25

Example: PSD of QAM with Rectangle Pulse Shaping

Baseband equivalent of QAM sequence

g t ak jbk pt kTs
k

ak , bk A,3 A

Autocorrelation of data sequence

Ra n E ak jbk ak n jbk n

E ak jbk 2 ,
n 0 10 A2 , n 0

n0
E ak jbk ak n jbk n , n 0
0,

PSD of baseband QAM sequence

g f 10 A2Tsinc 2 fT

Note that PSD of QAM has the

same general form as BPSK.

PSD of bandpass QAM sequence

s f

10 2
10
A Ts sinc2 f f c Ts A2Ts sinc2 f f c Ts
4
4
ECE414 Wireless Communications, University of Waterloo, Winter 2012

26

Some Practical Pulse Shapes

Below are some pulse shapes commonly used in communication systems:
Half-Sinusoid Pulse

t
pt A sin , 0 t T
T
AT jfT
1
1

P f
e
sinc fT - sinc fT
2

Full-Cosine Pulse
pt

A
t
1

cos
0 t T

2
T

P f

AT jfT
2sinc fT sinc fT 1 sinc fT 1
e
4
ECE414 Wireless Communications, University of Waterloo, Winter 2012

27

Some Practical Pulse Shapes (contd)

Gaussian Pulse
2 2 T 2
pt A exp
B t
ln 2 2

ln 2 f 2
A jfT ln 2

P f e
exp

2 B
T
2

where B is defined as the 3dB bandwidth of pulse

Raised Cosine Pulse
sin t T cost T

pt
2 2
2

t T

1 4 t T

0 1

: Roll-off factor

T
,
0

2T

1
T
1 1
T
P f 1 cos f
,

2
T
2T

2T

1
0
,
f

2T
ECE414 Wireless Communications, University of Waterloo, Winter 2012

28

Comparison of Pulse Shapes

Time-Domain

Square

Full-cosine
Half-sinusoid

Gaussian

ECE414 Wireless Communications, University of Waterloo, Winter 2012

29

Comparison of Pulse Shapes (contd)

Frequency-Domain
Full-cosine
Half-sinusoid
Gaussian
Square

Square
BW=2/T
Half-sinusoid
BW=3/T
Full-cosine
BW=4/T

2/T
3/T
4/T
ECE414 Wireless Communications, University of Waterloo, Winter 2012

30

Comparison of Pulse Shapes (contd)

Raised Cosine

0 1

: Roll-off factor

BW

1
T

1
2
BW
T
T
1/T
2/T
ECE414 Wireless Communications, University of Waterloo, Winter 2012

31

Optimum Receiver for AWGN

For a given SNR (i.e. a given signal power for fixed noise power), we aim to
achieve a low error probability. To calculate error probability, first we need to
identify the receiver structure.
The receiver consists of a demodulator and a detector:
The demodulator converts the received waveform r(t) into a N dimensional
vector r r1 , r2 ,...rN where N is the dimension of the signal-space for the
given modulation type.
The detector decides which of the possible M signal waveforms was
transmitted based on r, where M is the constellation size.
sm t

r t

nt

r
Demodulator

Detector

sm

r t sm t nt r s m n

ECE414 Wireless Communications, University of Waterloo, Winter 2012

32

Optimum Receiver for AWGN (contd)

Correlation-type demodulator

Matched-filter demodulator

For details, see Proakis Digital Communications Chapter 5

ECE414 Wireless Communications, University of Waterloo, Winter 2012

33

Optimum Receiver for AWGN (contd)

We want to design a signal detector that makes a decision based on the
observation of the vector r such that the probability of a correct decision is
maximized. The optimal decision rule is based on the maximization of so-called
a posteriori probabilities

ps m r : The probability of choosing sm m=1,2M based on the observation of r

This decision criterion is called the Maximum A Posteriori Probability (MAP) rule.
max ps m r

m1, 2...M

max

m1, 2...M

pr s m ps m
pr

max pr s m ps m
m1, 2...M

max pr s m

Bayes Theorem
p r : Common for all
ps m 1 M, i.e. Equally probable
messages

m1, 2...M

The conditional pdf pr s m is called the likelihood function and the decision
criterion based on the maximization of pr s m over the M signals is called the
maximum likelihood (ML) criterion.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

34

Optimum Receiver for AWGN (contd)

r sm n

For an AWGN channel, the components of the noise vector n are zeromean Gaussian random variables with variance N0/2

nk2
f nk
exp 2
2
2
2 2 N0
1

nk2
1

exp

N 0
N0

k 1,2...N

The received signal will have a Gaussian conditional distribution

pr s m p rk sm,k f rk sm,k
N

k 1

k 1

1
1
2
rk sm,k

exp
k 1 N 0
N0

N 0 N

1 N
2

exp
rk sm,k
2
N 0 k 1

ECE414 Wireless Communications, University of Waterloo, Winter 2012

35

Optimum Receiver for AWGN (contd)

The ML rule is then given as
max p r s m min rk sm,k 2 min r s m
N

m k 1

Distance metrics

The ML receiver decides in favor of the signal which is closest in Euclidean

distance to the received vector, r.
Expanding the decision rule,
min r s m

min r 2 r s m s m
m

where Em s m is the signal energy. Neglecting terms which do not affect

the decision and under the assumption that constant-energy modulation set
(e.g. PSK) is used
min r s m
m

max r s m

Correlation metrics

ECE414 Wireless Communications, University of Waterloo, Winter 2012

36

Example: Error Probability for BPSK

b 0 s1 t A cos2f ct 1 where 1 0
b 1 s2 t A cos2f ct 2 where 2

s1t A cos2f ct 2E T cos2f ct

s2 t s1 t

s2 t A cos2f ct 2E T cos2f ct

i.e. antipodal signaling

Unlike other M-PSK for M>2, we can represent this special form of BPSK
signal as 1-dimensional signal. The basis function is given as

2 T cos2f ct , 0 t T

1 t

0,

otherwise

Therefore, the optimal receiver has the following form of

r t

.dt

1 t

Euclidean
Distance
Decoder

ECE414 Wireless Communications, University of Waterloo, Winter 2012

37

Example: Error Probability for BPSK (contd)

Assume s1(t) is sent. Under the assumption of AWGN, the received signal
r t s1 t wt

The output of demodulator

T

r t 1 t dt s1 t wt 1 t dt E n

where
def T

n wt 1 t dt ~ N 0, N 0 2
0

Assume s2(t) is sent. The output of demodulator is now

T

r t 1 t dt E n

ECE414 Wireless Communications, University of Waterloo, Winter 2012

38

Example: Error Probability for BPSK

Decision regions
b 1
b 0

b0 E
b 1 E

r 0
1

Here we have two possible alternatives, therefore we can use a zero threshold
detector as an optimal detector.
Let P(e) denote the error probability

Pe P b 0, b 1 P b 1, b 0
P b 0 b 1 Pb 1 P b 1b 0 Pb 0

Bayes Theorem

Pb 0 Pb 1 1 / 2
P b 0 b 1 P b 1b 0

Pe P b 0 b 1

Equally probable messages

Due to symmetry

Under the assumption that b=1 is sent r E z

P b 0 b 1 Pr 0 b 1 f r b 1dr
0
ECE414 Wireless Communications, University of Waterloo, Winter 2012

39

Example: Error Probability for BPSK (contd)

P b 0 b 1 f r b 1dr
0

2
1 r E

dr
exp
N 0 0
N 0

y2
1

exp dy
2 2 E N0
2

r E
N0 2

2E

Q
N0

1 y2 2
where Q-function is defined as Q x
dy
e
2 x
40

Example: Error Probability for QPSK

00 s1 t A cos2f ct

01 s2 t A cos2f c t 2

10 s3 t A cos2f ct 3 2
11 s4 t A cos2f ct 2

2 Ts cos2f ct , 0 t Ts

1 t

0,

otherwise

2 Ts sin 2f ct , 0 t Ts

2 t

0,

otherwise

.dt

r t

1 t

Detector
s min r sm

m1, 2,3, 4

.dt

2 t

ECE414 Wireless Communications, University of Waterloo, Winter 2012

41

Example: Error Probability for QPSK (contd)

Under the assumption of AWGN (which exhibits symmetry), rotating and
moving the signal constellation does not change the error probability. Therefore,
we can rotate/move our signal constellation in such a way that the resulting
constellation allows easy mathematical derivation.
Here, we move our constellation as the target signal is located on the origin.
If there is no symmetry in the signal constellation, this should be repeated for
each signal.

Decision
regions

First, we calculate P(c), i.e. the probability of making a correct decision. Then,
probability of error is simply found as P(e)=1-P(c).
ECE414 Wireless Communications, University of Waterloo, Winter 2012

42

Example: Error Probability for QPSK (contd)

Assume that the signal located at the origin has been transmitted. If the
received signal is in the shaded area, this means we will make a correct decision.

d 2 Es

d 2

def T

nI wt 1 t dt ~ N 0, N 0 2
0
def T

nQ wt 2 t dt ~ N 0, N 0 2

d 2

Pc s1 P nI d 2, nQ d 2

P nI d 2 P nQ d 2

Es

Q
N0

Es

1 Q
N 0

P n Q

n ~ N , 2
2

Q 1 Q

ECE414 Wireless Communications, University of Waterloo, Winter 2012

43

Example: Error Probability for QPSK (contd)

Due to symmetry,

Pc Pc s1 Pc s2 Pc s3 Pc s4

Pe 1 Pc 2Q

2Q

Es Es
Q

N 0 N 0

Es 2 Eb

2 Eb 2 Eb
Q

N 0 N 0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

44

Example: Error Probability for BFSK

1

0 s1 t A cos 2 f c
t
2T

1

1 s2 t A cos 2 f c t
T

.dt

r t

1 t

Detector
s min r sm

m1, 2,3, 4

.dt

2 t
2 Ts cos2 f c 1 2T t , 0 t Ts
1 t
otherwise
0,
2 Ts cos2 f c 1 T t , 0 t Ts
2 t
otherwise
0,
ECE414 Wireless Communications, University of Waterloo, Winter 2012

45

Example: Error Probability for BFSK (contd)

2 t

d 2 Es

Es
Es

By rotation, it can be easily shown that

Es

Pe Q
N0

1 t

Now, we will study the same problem without rotation:

Assume s1 t s1 Es ,0 was sent. The received signal is r

Decision is based on min r s m
m

Es nI , nQ

max r s m
m

Es

Pe s1 Pr s 2 r s1 PnQ nI Es Pn Es Q
N0

nQ , nI ~ N 0, N0 2

def

n nQ nI ~ N 0, N 0

Es

Due to symmetry, Pe Pe s1 Q
N0
ECE414 Wireless Communications, University of Waterloo, Winter 2012

46

A Union Bound on Error Probability

In most cases, probability of error can not be obtained in closed form.
Therefore, one needs to find some bounds or approximations which can work for
any signal constellation.
We have already shown that the optimal decoder for any signal constellation
over AWGN is given by the Euclidean distance decoder, i.e. d m r s m 2
1 M
Pe Pe s m Ps m
Pe s m
M m1
m1
M

Pe s m P d l d m s m
l 1

l m

Pd l d m s m
M

l 1
l l

Ps m s l s m
M

Pe s m : Probability of making a
decision error when sm was sent
Union Bound (U-B)

P Ai P Ai
i i
Ps m s l : The probability of choosing sl
instead of the originally transmitted sm

l 1
l l

ECE414 Wireless Communications, University of Waterloo, Winter 2012

47

A Union Bound on Error Probability (contd)

d
Pe s m Ps m sl s m Q l ,m
l 1
l 1 2
M

l m

l m

where d l ,m s l s m

N
2 0
2

d l ,m

Q
l 1 2 N 0
l m

U B M

U-B: Union Bound

UB-B: UnionBhattacharyya Bound

UB B M

Q x e

x2 / 2

dl2,m
4 N0

l 1

l m

Assuming equal-probable message signals, the probability of error is

1 M
1 M M d l ,m

Pe
Pe s m
Q
M m1
M m1 l 1 2 N 0
l m

ECE414 Wireless Communications, University of Waterloo, Winter 2012

48

A Union Bound on Error Probability (contd)

The U-B requires the computation of all distances dl,m among signals in
the constellation. A looser bound can be obtained as follows

Pe s m Q l ,m M 1Q min
2 N0
l 1 2 N 0

l m
U B M

Minimum Euclidean
distance bound

where d min min dl ,m is the minimum Euclidean distance of the constellation.

l ,m

Then the probability of error is found as

1 M
1 M M dl ,m M 1Q d min

Pe
Pe s m
Q
2N
0

M m1
M m1 l 1 2 N 0
l m
P(e) is dominated by the minimum Euclidean distance of the signal
constellation.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

49

An Approximation for Error Probability

As an alternative, we can also the following approximate upper bound
Approximate upper bound

d l ,m ~
d min

Q
N dmin,m Q

2 N0
l 1 2 N 0

l m

U B M

Pe s m

N dmin,m : Number of signals at distance dmin from sm

d min
1 M
1 M
~

Pe
N dmin,m Q
P e s m
M m1
M m1
2 N0
d min

N dmin Q
2 N0

N dmin

1 M

N dmin ,m
M m1

ECE414 Wireless Communications, University of Waterloo, Winter 2012

50

Error Probability for M-PSK

Es

2
d min
4 Es sin 2 4 Eb log2 M sin 2
M
M
N dmin

1, M 2

2, M 2

2
Replacing d min
and N dmin into the formula on p.50, we obtain

2 Eb
,
M 2
Q
N
0

Pe
2Q 2 Eb log M sin 2 , M 2

2
N
M
0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

51

Error Probability for M-PSK (contd)

Error rate degrades as M increases.

Recall that spectral efficiency increases as M increases.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

52

Error Probability for QAM

sm t Am,r cos2f ct Am,i sin2f ct sm Am,r Ts 2, Am,i Ts 2
Am,r , Am,i A,3 A
Es0 Es3 Es12 Es15 9 A2Ts
s0

s1

s2

s3

Es5 Es6 Es9 Es10 A2Ts

s4

s5

s6

s7

Es1 Es2 Es4 Es8

s8

s9

s10

s11

s12

s13

s14

s15

Es7 Es11 Es13 Es14 5 A2Ts

Esavg

d min

d min 2 A Ts 2

2
Esavg
5

Es 4 Eb

1 M
2

Esm 5 A Ts
M m1

8
Ebavg
5

ECE414 Wireless Communications, University of Waterloo, Winter 2012

53

Error Probability for QAM (contd)

3 neighbours

2 neighbours
s0 , s3 , s12 , s15

4 neighbours

s0

s1

s2

s3

2 neighbours

s4

s5

s6

s7

3 neighbours

s8

s9

s10

s11

s12

s13

s14

s15

3 neighbours
s1 , s2 , s4 , s8 , s13 , s14 , s7 , s11
4 neighbours
s5 , s6 , s9 , s10
N dmin

1 M

N dmin, m 3
M m1

Using the result from p.50, we obtain an approximate upper bound

4 Ebavg
d min
~
3Q
Pe N dmin Q
5 N0
2N0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

54

Error Probability for QAM (contd)

Power efficiency decreases with increasing M, but not early as fast as M-PSK.

Recall that spectral efficiency increases as M increases.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

55

Error Probability for M-FSK

sm t A cos2 f c f m t

f m m 2T , m 1,2...M

0 t Ts

Each signal occupies its own

dimension. Therefore, each
signal has M-1neighbours,
separated from each other by

d min 2 Es

Es
Eb log2 M

Pe M 1Q
M 1Q

N0
N0

ECE414 Wireless Communications, University of Waterloo, Winter 2012

56

Error Probability for M-FSK (contd)

BPSK
2 Eb

Pe Q
N0

BFSK
Eb

Pe Q
N0

As M increases, power efficiency improves (i.e. less Eb is required).

Recall that spectral efficiency decreases as M increases.

For M=2, BFSK requires 3dB more energy/bit to achieve the same P(e) as BPSK.
In other words, BPSK is 3dB more power efficient that BFSK.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

57

Comparison of Power Efficiency of Modulation Schemes

We will use BPSK/QPSK as a benchmark with which to compare the
power efficiency of other modulation schemes.
2
BPSK/QPSK has d min
4 Eb . Now, define the power efficiency of a
modulation scheme (relative to BPSK/QPSK) as
2
d min
P 10 log10
4 Eb

ECE414 Wireless Communications, University of Waterloo, Winter 2012

58

Differential Phase Shift Keying (DPSK)

So far, we assumed that coherent demodulation is performed, i.e. that the
carrier phase is perfectly known at the receiver. This normally requires carrier
phase estimation.
An alternative is differentially encoding, where the data is encoded in
phase difference from one symbol to the next. Assuming binary signalling,
d k 0,1 bk 1,1 ak ak 1bk

dk

bk

+1

-1

-1

-1

+1

-1

+1

-1

+1

-1

-1

+1

ak

+1

This diagram might

correspond to either
PSK or DPSK!

ECE414 Wireless Communications, University of Waterloo, Winter 2012

59

Transmitter and Receiver for DPSK

d k 0,1

Mapper

bk 1

Differential
Encoder

a k 1

s DPSK t

A cosc t

represents any mismatch between transmitter/receiver oscillators or phase

introduced by the channel. In our system model, (independent of where it comes
from) we included in the transmitter block.
In coherent systems, we need to estimate and compensate this phase error at
the receiver. Here, we simply ignore it!
y I t
dt
r t

t kT

2 cos c t

yQ t
2 sin c t

dt

yk

1 Symbol
Delay

zk

sgn Re
.

y k* 1

ECE414 Wireless Communications, University of Waterloo, Winter 2012

60

Error Probability of DPSK

st Aa k cosct

k 1T t kT

yt yI t jyQ t
ak Ae j N t
y k yt dt Ae j ak dt N t dt ATe j ak N k
T

N k ~ N 0,2 N 0T

def

def

Defining 1 y k y k 1 2 2 y k y k 1 2 the decision variable

can be written as

2 2

Pe Pbk 1bk 1 PReyk yk*1 0 bk 1 P1 2 bk 1
z k sgn Re y k y k* 1 1 2

ECE414 Wireless Communications, University of Waterloo, Winter 2012

61

Error Probability of DPSK (contd)

We need to find statistical properties of 1 and 2 :
First, we recall the definitions
def

def

1 y k y k 1 2 2 y k y k 1 2

1 ATe j ak ak 1 / 2 N k N k 1 2
2 ATe j ak ak 1 / 2 N k N k 1 2
E1 ATe j ak ak 1 / 2
E2 ATe j ak ak 1 / 2

2
1

Var1 E 1 E1 E N k N k 1 N k N k 1 * N 0T
4

2
1

Var2 E 2 E2 E N k N k 1 N k N k 1 * N 0T
4

ECE414 Wireless Communications, University of Waterloo, Winter 2012

62

Error Probability of DPSK (contd)

Encoding scheme: ak ak 1bk
ak

ak 1

bk

ak ak 1 2

ak ak 1 2

+1

+1

+1

+1

-1

+1

-1

-1

-1

-1

+1

-1

+1

-1

-1

Under the assumption that bk 1 is sent

E1 0

1R , 1I ~ N 0, 2

Var 1 N 0T

1 1R j1I

Complex Gaussian

E2 ATe j

2
2 R ~ N AT cos , 2 2 I ~ N AT sin ,
2 2 R j2 I Complex Gaussian

Var 2 N 0T

where 2 N 0T 2

1 : Rayleigh, 2 : Rician
ECE414 Wireless Communications, University of Waterloo, Winter 2012

63

Error Probability of DPSK (contd)

1 : Rayleigh
2 : Rician

1
1
f 1 2 exp 2
2

2
1
2

f 2 2 exp 2 2 2
2

2
I 0 2

where the non-zero mean is found as 2 A2T 2 cos2 sin 2 AT

Now, we return to P(e) computation

Pe P1 2 P1 2 f 2 d
0

1
2

1
P 1 2
exp 2 d 1

exp u du

2 2 2

exp 2 2 2

def

1 : Rayleigh
2 : Rician
1
2 2

ECE414 Wireless Communications, University of Waterloo, Winter 2012

64

Error Probability of DPSK (contd)

2
2 2
I
Pe exp 2 2 exp
d
2 0 2
2
Variable change
0
2

def
2
2

x x
1 x
x 2

2 exp
I
dx

2 0
2
2 0
2

def
m 2
m2 x
x 2 m 2 mx
1
I
exp 2 2 exp
dx
AT
2 0 2
2
2
2 0

A T
1

exp

2
2N0

=1

E
1

exp
2
N0

2 N 0T 2

E s 2 t dt A2T 2
T

ECE414 Wireless Communications, University of Waterloo, Winter 2012

65

Error Probability of DPSK (contd)

Coherent vs. Differential PSK

10

Coherent
Differential
-1

10

-2

BER

10

-3

10

-4

10

-5

10

-6

10

6
SNR [dB]

10

12

There is some performance degradation due to differential detection,

but now a less complex receiver can be used (i.e. no need for phase
tracking).

ECE414 Wireless Communications, University of Waterloo, Winter 2012

66

Temporal Characteristics of Modulation Schemes

So far, we have considered pulse shapes which are strictly limited in the
symbol interval. By using a pulse shape to spill over into adjacent symbol
intervals, better spectral efficiency can be achieved, however this also results in
intersymbol interference (ISI).
The following block diagram is commonly used for studying ISI. Assuming
matched filter type implementation for the demodulator,
t lTs

g t ak p t kTs
k

hT t

hC t

hR t

Detector

wt
Actual Channel

Equivalent Channel

zl ak heq l k Ts nlTs where heq hT t hC t hR t

k
nt wt hR t
ECE414 Wireless Communications, University of Waterloo, Winter 2012

67

Temporal Characteristics of Modulation Schemes (contd)

Here, we use in pulse shapes which spill over adjacent symbols. This will
bring ISI terms:

zl ak heq l k Ts nlTs al heq 0 ak heq l k Ts nlTs

k l

The condition for no ISI is

ISI terms

1, n 0
heq nTs
0, otherwise

In frequency domain, this requires

l
H eq f constant
Ts
l

See proof Proakis Digital Communications Chapter 9

This condition is known as Nyquist pulse-shaping criterion or Nyquist
condition for zero ISI.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

68

Temporal Characteristics of Modulation Schemes (contd)

In the following, we consider three distinct cases:
Ts

1
2W

l
H eq f
Ts
l

W: Bandwidth of equivalent ch.

For this case, there is no choice for Heq to satisfy Nyquist criterion.

Ts

1
2W

l
H eq f
Ts
l

For this case, there is only one solution:

Ts , f W

H eq f

0, otherwise

ECE414 Wireless Communications, University of Waterloo, Winter 2012

69

Temporal Characteristics of Modulation Schemes (contd)

Ts

1
2W

l
H eq f
Ts
l

For this case, there exists many solutions as to satisfy H eq f cons.

Ts
l

A particular pulse shape which satisfies the above property and has been widely
used in practical applications is raised cosine. (See page 28) The Nyquist
pulse takes zero at the sampling points for adjacent signalling intervals.

X f l Ts cons.
l

ECE414 Wireless Communications, University of Waterloo, Winter 2012

70

Temporal Characteristics of Modulation Schemes (contd)

Under the matched-filter assumption (i.e. which maximizes the output
signal-to-noise ratio), the transmit and receive filters satisfy

HT f H R f
Under the ideal channel assumption , i.e. H C f 1

HT f H R f

H eq f

For raised-cosine equivalent channel response, we can divide it into two

root-raised-cosine (RRC) filters.

Ts ,

Ts
Ts
1

f
,
H RRC f H RC f cos
2Ts

2

0,

1
2Ts

1
1
f
2Ts
2Ts
f

1
2Ts

ECE414 Wireless Communications, University of Waterloo, Winter 2012

71

Temporal Characteristics of BPSK

Consider the baseband BPSK modulated signal with RRC pulse shape

g t ak p t kTs
k

ak 1

Eye pattern is a sketch of g(t) for all possible combinations of a1 , a2 , a3 ,...

Minimum instantaneous
power=0

Maximum instantaneous
power=(1.6)2=4.1 [dB]
Dynamic range=4.1 [dB]
Average power=1=0 [dB]
Peak power Avg. power 4.1dB
For this example, we observe large dynamic range of instantaneous power
and large peak/average ratio. These make the design of TX power amplifier
difficult.
ECE414 Wireless Communications, University of Waterloo, Winter 2012
72

Temporal Characteristics of QPSK

The QPSK signal with pulse shaping can be written as

g t ak pt kTs cosc t bk pt kTs sin c t

ak , bk 1

The instantaneous power of the QPSK signal is

2

ak pt kTs bk pt kTs
k
k

Hence, a QPSK signal suffers similar time-domain problems as a BPSK

signal. Now assume, different pulses are used for I&Q channels. If Q channel
pulse is delayed by 1/2 symbol relative to I channel pulse, i.e. qt pt Ts 2
the instantaneous power is
2
2

ak pt kTs bk pt kTs Ts 2
k
k

Both terms can not pass through zero simultaneously, hence significantly
increasing the minimum instantaneous power and reducing dynamic range of the
signal. PSD and BER remain unchanged. This is known as Offset QPSK (OQPSK).
ECE414 Wireless Communications, University of Waterloo, Winter 2012

73

Temporal Characteristics of QPSK (contd)

Another variant of QPSK is /4-QPSK. This modulation scheme is a
superposition of two QPSK signal constellations offset by /4 relative to each
other.

ak jbk

e j 2 ,e j

ak jbk

e j 4 ,e j3 4

ak jbk

e j 4 , e j3 4 , for even k

e j 2 , e j , for odd k

PSD and BER of /4-QPSK are the same as QPSK.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

74

Temporal Characteristics of QPSK (contd)

In QPSK, transitions between opposite points in the signal constellation

cause the instantaneous power to zero, leading to a large dynamic range.
The special structure of /4-QPSK avoids transitions which pass the origin,
reducing dynamic range and peak-to-average power ratio.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

75

Continuous FSK
We can get perfect temporal properties by using continuous FSK (CFSK)

g t ak pt kTs

ak 1

st cos2f ct 2h g d cos2f ct 2h ak qt kTs

where

def t

qt p kTs d

h: Modulation index

Instantaneous power= constant

Dynamic range=0dB
Peak-to-average power ratio=0dB
There is no abrupt switching from one phase to another, avoiding phase
discontinuities.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

76

Continuous FSK (contd)

Here, we assume a rectangle pulse shape for p(t).
p(t): Frequency pulse

t ; a 2h ak qt kTs nTs t n 1Ts

k

n 1

h ak 2h

1/2Ts

k 0

t nTs
an
2Ts

3h
t
Ts

q(t): Phase pulse

2h
h
0
-h
-2h
-3h

1/2

t
Ts

Phase Tree

+1
+1

n=0,1,..

-1
+1

-1

-1

Ts

2Ts

3Ts

The shaded path illustrates the phases for

the input sequence {+1,+1,-1}

ECE414 Wireless Communications, University of Waterloo, Winter 2012

77

Continuous FSK (contd)

n1

t; a 2h ak 2h
k 0

t nTs
ha
an 2 n t n hnan
2Ts
2Ts

nTs t n 1Ts

han
t n hnan
st cos2f ct t ; a cos2 f c
2Ts

h
t n hn
st cos2 f c
2Ts

h
t n hn
an 1 st cos2 f c
2Ts

an 1

The separation between two carriers is f h T s

For orthogonality, the minimum value for h should be chosen as h=1/2. This
special case is known as Minimum Shift Keying (MSK).
ECE414 Wireless Communications, University of Waterloo, Winter 2012

78

Continuous FSK (contd)

We have already introduced MSK as a special case of modulation family of
CFSK.
An MSK signal can be also considered as a special form of OQPSK where
the rectangular pulses are replaced with half-sinusoidal pulses.

st ak pt 2kTs cos2f ct ak pt 2kTs Ts sin 2f ct

odd k

even k

t
,0 t 2Ts
cos
pt 2Ts
0,
otherwise

The transmission rate on the two orthogonal carriers is 1/2Ts bits/sec so that
the combined transmission rate is 1/Ts bits/sec.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

79

Comparison of MSK, QPSK and OQPSK

Continuous phase is assured in MSK while 90 and 180 phase changes are
observable for OQPSK and QPSK respectively.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

80

Comparison of MSK, QPSK and OQPSK

(contd)
In terms of temporal
efficiency, MSK obviously
outperforms QPSK and
OQPSK.
The main lobe of MSK is
wider than that of QPSK and
OQPSK and, in terms of null-tonull bandwidth MSK is less
spectral efficient.
MSK has lower sidelobes
than QPSK and OQPSK Less
adjacent channel interference
MSK, QPSK and OQPSK
have the same power efficiency.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

81

Gaussian MSK
The spectral efficiency of MSK can be further improved by prefiltering.

g t ak pt kTs
k

Gaussian
LPF

MSK
Modulator

The frequency response function of Gaussian LPF filter is given as

f 2 ln 2
2 2 B 2t 2
2
ht

H f exp
B exp

ln 2
ln 2
B 2

where B is 3dB-bandwidth of the filter.

We are interested in how a rectangle pulse passed through a Gaussian
LPF will look like.

ECE414 Wireless Communications, University of Waterloo, Winter 2012

82

Gaussian MSK (contd)

Frequency pulse
t T / 2
t T / 2
f t Q

T
T

2BT

ln2

BT: Normalized
3dB-Bandwidth

Phase pulse

x
1 1
qt 1
exp
2
2

x1

t T / 2
T

x2

x2

xQ x

x1

t T /2
T

Phase pulse corresponding to rectangular

pulse shaping (i.e. no filtering) is also
included in the figure.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

83

Gaussian MSK (contd)

BT: Normalized
3dB-Bandwidth
of Gaussian filter

For BT , the pulse shape takes its original unfiltered form , i.e. rectangle
pulse. GMSKMSK
The frequency pulse has a duration of 2Ts although signaling rate is 1/Ts. Such a
LPF will result in intersymbol interference which requires sequence estimation for
optimal detection.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

84

Gaussian MSK (contd)

BT should be chosen as to find a good
compromise between spectral efficiency
and ISI.
As BT decreases, the spectral
efficiency improves (i.e. less bandwith).
Also sidelobes fall off very rapidly (i.e.
less adjacent channel interference).
However, reducing BT results in ISI
and error rate performance degrades (i.e.
observation of an irreducible error floor
due to ISI)

In practical application, BT is typically

chosen as (0.2, 0.5). GSM systems use
GMSK with BT=0.35.
ECE414 Wireless Communications, University of Waterloo, Winter 2012

85