Chapter 6 Digital Modulation

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 43

Chapter 6: Digital Modulation 1

Chapter 6: Digital Modulation


Most real communication channels have very poor response in the
neighbourhood of zero frequency and hence can be regarded as
bandpass channel.
Baseband digital signals can have frequency component as low as DC,
so it will be grossly distorted if transmitted over a bandpass channel.
Digital modulation allows a baseband digital signal to be translated to
a higher frequency range centering around a carrier frequency.
The higher frequency range is within the passband of the bandpass
channel, hence allowing minimum attenuation transmission.
1 0 1 1
t t
1 0 1 1
1 0 1 1
x(t)
x
s
(t)
0
0
X(f)
f
f
X
s
(f)
Unipolar
NRZ
f
c
f
c
carrier freq f
c
Chapter 6: Digital Modulation 2
There are three basic Digital Modulation techniques :-
Amplitude Shift Keying (ASK)
Frequency Shift Keying (FSK)
Phase Shift Keying (PSK)
Modulation can be binary or Mary (more than two
levels).
Chapter 6: Digital Modulation 3
x(t)
c(t) = sin 2f
c
t
s(t)
x(t) is always a unipolar NRZ waveform for ASK.
0 T
b
0
V
0
-1
0
1
0
-V
0
V
2T
b
3T
b
4T
b
5T
b
1 0 1 1 0
c(t) = sin 2f
c
t
x(t)
T
b
2T
b
3T
b
4T
b
5T
b
s(t) BASK waveform
Generation of Binary Amplitude Shift Keying (BASK) signal
Chapter 6: Digital Modulation 4
Amplitude Spectrum of BASK
Assume x(t) is transmitting a long series of 10101010 .
Assume f
c
>> r
b
where r
b
is the bit rate of x(t).
|S(f)|
(Showing only the +ve freq)
4
V
Amplitude spectrum of BASK
0
0
f
2
V
2
r
b
2
r
b

2
r 3
b
2
r 3
b

|X(f)|
*
-f
c
0 f
c
f
|C(f)|
1/2
Chapter 6: Digital Modulation 5
Generation of a binary FSK (BFSK) signal involves generating two
BASK signals.
The first BASK signal x
ask2
(t) is obtained by the unipolar NRZ
waveform x(t) modulating a carrier frequency f
2
. (Upper branch).
The second BASK signal x
ask1
(t) is obtained by passing x(t) through

a NOT gate to obtain , and let modulates another
carrier frequency f
1
. (Lower branch).
s(t) = x
ask2
(t) + x
ask1
(t) is the required BFSK waveform.
Given a carrier frequency f
c
,
f
2
= f
c
+ f
d
; f
1
= f
c
f
d
;
where f
d
is called the frequency deviation.
Generation of Binary Frequency Shift Keying (BFSK) signal
x(t)
x
sin(2f
2
t)
x
sin(2f
1
t)
+ x(t)
s(t)
x(t)
x
ask2
(t)
x
ask1
(t)
x(t)
Chapter 6: Digital Modulation 6
x
sin(2f
2
t)
x
sin(2f
1
t)
+ x(t)
s(t)
x(t)
x
ask2
(t)
x
ask1
(t)
0 T
b
0
V
0
0
0
0
0
0
0
2T
b
3T
b
4T
b
5T
b
V
0
T
b
T
b
T
b
T
b
2T
b
2T
b
2T
b
2T
b
3T
b
3T
b
3T
b
3T
b
4T
b
4T
b
4T
b
4T
b
5T
b
5T
b
5T
b
5T
b
V
V
V
-V
-V
-V
x(t)
x
ask2
(t)
x
ask1
(t)
x(t)
s(t)
1 0 1 1 0
Chapter 6: Digital Modulation 7
Amplitude Spectrum of BFSK
Assume x(t) is transmitting a long series of 10101010 .
Assume f
c
>> r
b
where r
b
is the bit rate of x(t).
0
0
f
2
V
2
r
b
2
r
b

2
r 3
b
2
r 3
b

|X(f)|
f
2
-r
b
/2
|S(f)|
(Showing only the
+ve freq)
4
V
Amplitude spectrum of
BFSK
f
1

0
f
f
1
+r
b
/2
f
2

f
1
-r
b
/2
f
2
+r
b
/2
f
2
-3r
b
/2
f
1
+3r
b
/2
0
0
f
2
V
2
r
b
2
r
b

2
r 3
b
2
r 3
b

| X(f) |
*
-f
2
-f
1
0 f
1
f
2
f
|C(f)|
1/2
*
Chapter 6: Digital Modulation 8
x(t)
c(t) = sin 2f
c
t
s(t)
x(t) is always a polar NRZ waveform for PSK.
Generation of Binary Phase Shift Keying (BPSK) signal
0 T
b
2T
b
3T
b
4T
b
5T
b
1 0 1 1 0
c(t) = sin 2f
c
t
x(t)
s(t) BPSK waveform
0
V
0
-1
0
1
0
0 T
b
2T
b
3T
b
4T
b
5T
b
V
-V
-V
Chapter 6: Digital Modulation 9
Amplitude Spectrum of BPSK
Assume x(t) is transmitting a long series of 10101010 .
Assume f
c
>> r
b
where r
b
is the bit rate of x(t).
|S(f)|
(Showing only the +ve freq)
Amplitude spectrum of BPSK
0
0
f
2
r
b
2
r
b

2
r 3
b
2
r 3
b

|X(f)|
*
-f
c
0 f
c
f
|C(f)|
1/2
V
2
V
Chapter 6: Digital Modulation 10
Optimal Receiver for Binary Digital Modulation Systems
The diagram below shows the overall diagram of a binary data
transmission over a bandpass channel.
The source is generating a binary bit stream b
k
, and represented
electrically by a line code.
The line code waveform modulates a carrier to produce a digitally
modulated signal (e.g. ASK / FSK / PSK) which is transmitted
across the bandpass channel.
Ideally, the demodulator reproduces a replica of the b
k
bit stream, but
because of the additive white Gaussian noise (AWGN) introduced at
the channel, some of the bits will be different (bit error).
The function of the matched filter is to minimise these errors.
Modulator
Bandpass
Channel
Threshold
Device
Sample
every T
b
sec
Demodulator
Matched
Filter
Carrier Clock
Carrier
Clock
k
b
k
b

AWGN
Chapter 6: Digital Modulation 11
How to Design a Matched Filter for Digital Modulation Signal ?
Example 1.1
Design a matched filter for the case
when its input is an ASK waveform as
shown below.
Solution:
Input
waveform
of matched
filter
0
V
-V
1 0 1 1 0
Basic waveforms:
s
2
(t)
0
V
-V
s
1
(t)
0
0
V
-V
0
V
-V
s
2
(T
b
-t) s
1
(T
b
-t)
s
2
(t)
s
1
(t)
s
2
(t) s
1
(t)
s
2
(-t) s
1
(-t)
{
Impulse response h(t)
0 T
b

0
V
-V
0
V
0
-V
Chapter 6: Digital Modulation 12
Therefore if the output of the bandpass channel feeding the
matched filter is a BASK waveform of amplitude V volt,
then the matched filter must have a characteristic where its
impulse response is as shown:
Matched
Filter
(t)
0 t
Impulse
response
h(t)
Once you know the impulse response characteristic of the
matched filter, you can implement the matched filter circuit
from the h(t) waveform using signal processing technique.
V
0
-V
T
b
t

Chapter 6: Digital Modulation 13
Coherent BPSK System
Matched Filter
b
k
is a polar NRZ waveform.
The bandpass filter (BPF) limits the noise power entering the matched
filter.
The matched filter is implemented by the Integrate and Dump
Correlation receiver.
Chapter 6: Digital Modulation 14
Example 1.2
An integrate and dump correlation receiver is shown in Fig 1.2.1.
If its input is a BPSK waveform of amplitude V volt and f
c
= 2 r
b
,
where f
c
is the carrier frequency and r
b
the bit rate, sketch the
waveforms at A to E for a 1101 sequence. Explain the operations
of SW1 and SW2. Assume an ideal noiseless bandpass channel
with infinite bandwidth. Assume r
b
= 1200 b/s.
Chapter 6: Digital Modulation 15
Solution:
For binary sequence { 1 1 0 1}, BPSK waveform at A is:
1 1 0 1
t
-V
V
0
Chapter 6: Digital Modulation 16
Waveform at B is based on the equation: s
2
(t) s
1
(t) for each bit frame:
To sketch waveform B, repeat
s
2
(t) s
1
(t) waveform pattern for every
bit frame.
s
2
(t) (binary 1)
s
1
(t) (binary 0)
s
2
(t) s
1
(t) :
V
-V
2V
Bit
Duration
V
-V
-2V
2V
t
1 1 0 1
Waveform B :
-2V
Chapter 6: Digital Modulation 17
Waveform C is the multiplication of Waveform A and Waveform B
A:
B:
-2V
2
2V
2
t
C:
1 1 0 1
-V
V
t
0
2V
t
-2V
0
0
Consider binary 1 frame:
t V V
t
V
t V
c
c
c
e
e
e
2 cos
2
2 cos 1
2
sin 2 B A x
2 2
2
2 2
=

=
=
Consider binary 0 frame:
t V V
t
V
t V
c
c
c
e
e
e
2 cos
2
2 cos 1
2
sin 2 B A x
2 2
2
2 2
+ =

=
=
Chapter 6: Digital Modulation 18
Waveform D is the integration of Waveform C.
Consider an integrator circuit as shown in Fig (a) below during binary 1 frame:
) (t v
o
D =

D varies
linearly with t
over 0 t
T
b

Value
of D at
t = T
b
Fig (a)
C = t V V t v
c i
e 2 cos ) (
2 2
=
K = circuit constant
D =
| | | |
| | | | | |
b
T T
c
T
T
c
c
T
T
c
T
o
T KV t KV t KV t KV
t
KV
t KV dt t V K dt V K t v
b b b
b b
b b
2
0
2
0
2 5 -
0
2
0
2
0
2
0
2
0
2
2 sin 10 x 3 . 3
2 sin
2
2 cos ) (
= ~ =
= =
} }
e
e
e
e

2V
2




-2V
2




KV
2
T
b



t
t
0
C
D
0
-KV
2
T
b



1 1 0 1
T
b
Chapter 6: Digital Modulation 19
SW1 is closed at the end of each bit duration for a very short duration to
sample the D waveform. After sampling D, SW1 is opened again followed
by the short closure of SW2 to discharge the capacitor so that D
waveform drops to zero to initialize the start of the next bit-frame
waveform of D.
t
t
D
= SW1 operation
= SW2 operation
1
0 1 1
0
E
KV
2
T
b
-KV
2
T
b
0
V
-V
Chapter 6: Digital Modulation 20
Waveform A to E:
A:
B:
-2V
2
2V
2
t
C:
1 1 0 1
-V
V
t
0
2V
t
-2V
0
0
t
t
D
= SW1 operation
= SW2 operation
1
0 1 1
0
E
KV
2
T
b
-KV
2
T
b
0
V
-V
Chapter 6: Digital Modulation 21
Example 1.3
The input to a matched filter from a bandpass transmission channel is
a BPSK signal as shown in Fig 1.3.1. Each bit has a duration of T
b

sec. The single-sided power spectral density of the channel is given
as q W/Hz. Show that the probability of bit error for this matched
filter is given by:
|
|
.
|

\
|
=
2
T V
erfc
2
1
P
b
2
e
Assume statistical independence in bit generation and
equiprobable transmission of binary 1 and 0.
1 1 0 1
t
-V
V
0
Fig 1.3.1
Chapter 6: Digital Modulation 22
Solution:
Probability of bit error for a matched filter in general is given by:
|
|
.
|

\
|
= |
.
|

\
|
=
8
erfc
2
1
2 2
erfc
2
1
P
e

where
}
=
b
T
dt t s t s
0
2
1 2
2
)] ( ) ( [
2
q

}
Given in Formula List
Doc
2
2
2
1
= = = =
c
b
b
b
c
c
T
T
T
r
T
f
Note:
From Example 1.2
s
2
(t) s
1
(t)
waveform is :-
2V
T
b
-2V
s
2
(t) s
1
(t) = 2V sin
c
t
T
c
| | | |
q
t
e q
t
e q
e
e q
e
q
e
q
e
q q

b
c
b
c
b
c
b
T
c
c
T
T T
c
T
c
T
c
T
T V
T
V
T
T
T
V
t t
V
dt t dt
V
dt
t V
dt t V dt t s t s
b b
b b b
b b
2 2
2
0 0
2
0 0
2
0
2
0
2 2
0
2
1 2
2
4
8 sin
2
1 4
4
sin
2
1 4
2 sin
2
1 4
2 cos
4
2
2 cos 1 8
sin 4
2
)] ( ) ( [
2
=
|
|
.
|

\
|
=
|
|
.
|

\
|
=
|
|
.
|

\
|
=
|
.
|

\
|
= |
.
|

\
|

=
= =
} } }
} }
Chapter 6: Digital Modulation 23
|
|
.
|

\
|
= |
.
|

\
|
=
8
erfc
2
1
2 2
erfc
2
1
P
e

(from Formula List)
Substitute
q

b
T V
2
4
=
into above equation:
|
|
.
|

\
|
=
|
|
.
|

\
|
=
|
|
.
|

\
|
=
q
q q
2
erfc
2
1
8
4
erfc
2
1
8 /
4
erfc
2
1
P
2
2 2
e
b
b b
T V
T V T V
i.e.
|
|
.
|

\
|
=
q 2
erfc
2
1
P
2
e
b
T V
for matched filter with BPSK input.
q

b b
T V T V
2 2
2
4 4
= =
Chapter 6: Digital Modulation 24
Differential Phase Shift Keying (DPSK)
In the coherent BPSK system mentioned earlier, it was
assumed that the carrier signal s
2
(t) s
1
(t) at the integrate
and dumped correlation receiver was synchronised with the
carrier of the incoming BPSK waveform in terms of frequency
and phase.
Also the two switches at the receiver that samples every T
b

second must synchronise with the bit frame of the BPSK
waveform.
Synchronisation of carrier frequency and bit clock requires
extra overhead signalling if transmitted together with the
transmission signal or if generated locally would require
complex hardware at the receiver.
A DPSK system gets around the need for a coherent reference
signal by encoding the bit stream at the source before generating
the BPSK signal. At the receiver, it uses the previous BPSK bit-frame
wave to correlate with the present BPSK bit-frame wave.
Chapter 6: Digital Modulation 25
DPSK System
Chapter 6: Digital Modulation 26
Example 1.4
Fig 1.4.1 shows the block diagram of a DPSK system. The binary
input is in unipolar NRZ format of amplitude 2V volt, at a bit rate
of 1200 b/s. The carrier is sin
c
t where
c
= 4800rad/s.
Assume that the input is a long series of 10101010 . Assume
the receiver bandpass filter (BPF) has negligible effect on the
shape of the received signal waveform. The receiver lowpass
filter is assumed ideal and has a cut-off frequency f
co
= r
b
/2 =
600 Hz. Sketch the waveforms at points A to G as indicated in Fig
1.4.1 for a 1010 frame. Assume distortionless transmission path.
Also assume that the encoder output is binary 1 prior to the 1010
frame.
Chapter 6: Digital Modulation 27
Fig 1.4.1
Transmitter
Receiver
Chapter 6: Digital Modulation 28
Solution:
A has unipolar format; bit rate r
b
= 1200 b/s
sec
2400
1
Period Carrier Hz 2400 4800 2
rad/s 4800 2
= = =
= =
c c c
c c
T f f
f t t e
sec
1200
1
=
b
T
B
delay
One bit duration
c
b
b
T
r
T 2
2400
1
x 2
1200
1 1
= = = =
= two carrier periods
Chapter 6: Digital Modulation 29
X Y Z(o/p)
0 0 1
0 1 0
1 0 0
1 1 1
X-NOR
B
delay
-T
b
0 T
b
0
2V
0
0
2V
0
0
2V
0
-V
0
V
0
-1
0
1
0
-V
0
V
-T
b
-T
b
-T
b
-T
b
-T
b
T
b
T
b
T
b
T
b
T
b
2T
b
2T
b
2T
b
2T
b
2T
b
2T
b
3T
b
3T
b
3T
b
3T
b
3T
b
3T
b
4T
b
4T
b
4T
b
4T
b
4T
b
4T
b
1 0 1 0
A
B
delay
B
C
sin
c
t
D
1
0
1
1 1
0
0 0
1
Transmitter:
Chapter 6: Digital Modulation 30
Receiver:
D
delay
If D and D
delay
has same polarity:
t
V V
t
V
t V
c
c
c
e
e
e
2 cos
2 2
2
2 cos 1
sin D x D E
2 2
2
2 2
delay
=

=
= =
If D and D
delay
has different
polarity:
t
V V
t
V
t V
c
c
c
e
e
e
2 cos
2 2
2
2 cos 1
sin D x D E
2 2
2
2 2
delay
+ =

=
= =
-T
b
0
-V
0
V
0
0
V
0
-V
V
2
-V
2
0
T
b
T
b
T
b
2T
b
2T
b
2T
b
3T
b
3T
b
3T
b
4T
b
4T
b
4T
b
D
D
delay
E
Chapter 6: Digital Modulation 31
0
0
V
2
0
0
K
-1
0
0
2V
-V
2
-K
T
b
2T
b
3T
b
4T
b
2
T
b
2
3T
b
2
5T
b
2
7T
b
2
T
b
2
3T
b
2
5T
b
2
7T
b
1 0 1 0
E
F
G
E waveform consists of a square
wave (cyan) and high freq (2 f
c
)
phase-switching ripple added to it.
E passes through a LPF of cut-off
freq r
b
/2 Hz which retains the
fundamental freq of the cyan
square wave, but remove all other
higher freqs.
Hence F is a sine wave of
freq = r
b
/2 = 1200/2 = 600 Hz .
F is sampled at its highest
excursion value (centre of each bit
interval).
If F > 0, G = binary 1 (2V volt)
else G = binary 0 ( 0 volt).
G bit stream is hence = A bit stream.
Chapter 6: Digital Modulation 32
Probability of bit error for DPSK is:
|
|
.
|

\
|
=
2
T V
exp
2
1
P
b
2
e
where q is the single-sided power spectral density of the
white channel noise. P
e
for DPSK is hence higher (worst) than
BPSK.
Disadvantages of DPSK:
Asynchronous transmission is not possible because of the
need to synchronise the previous bit frame with the current
bit frame at the receiver correlator to allow multiplication .
Another minor error is that an error will propagate to the
adjacent bit.
Example
Case 1: No error Case 2: Error
Received phase (D) 0 0 0 0 0 0 0
Correlation output (F) + + - + - - - - + -
Output sequence (G) 1 1 0 1 0 0 0 0 1 0
Red = Error; Green = Propagated error
Chapter 6: Digital Modulation 33
Quadrature Phase Shift Keying (QPSK)
In binary phase shift keying (BPSK):
Binary 0 carrier with 180 degree ( rad) phase shift.
Binary 1 carrier with 0 degree (0 rad) phase shift.
In quadrature phase shift keying (QPSK), 2 bits are used to form
4 possible phase shifts:
Binary-pair 00 carrier with 0 degree (0 rad) phase shift.
Binary-pair 01 carrier with 90 degree (/2 rad) phase shift.
Binary-pair 11 carrier with 180 degree (rad) phase shift.
Binary-pair 10 carrier with 270 degree (3/2 rad) phase shift.
Chapter 6: Digital Modulation 34
If the carrier frequency of the QPSK system is a cosine wave, the QPSK
modulated signal can be expressed as:

=
s s +
=
" 10 " 270
" 11 " 180
" 01 " 90
" 00 " 0
0 ) 2 cos(
) (
f or
f or
f or
f or
T t t f A
t s
i
b i c
i

|
| t
where each dibit (2 bits) forms a phase-shifted carrier wave
symbol.
Hence the symbol rate r
s
for QPSK is half the bit rate (r
b
);
i.e. r
s
= r
b
/2
Chapter 6: Digital Modulation 35
Example 1.5
A QPSK modulation scheme can be described by the equation below:

=
s s +
=
" 10 " 270
" 11 " 180
" 01 " 90
" 00 " 0
0 ) 2 cos(
) (
f or
f or
f or
f or
T t t f A
t s
i
b i c
i

|
| t
If the input to the modulator is a binary bit stream: 00011011
sketch the QPSK modulated waveform.
Solution:
Chapter 6: Digital Modulation 36
Probability of Bit Error (P
e
) for QPSK
Consider the QPSK coding scheme shown earlier:

=
s s +
=
" 10 " 270
" 11 " 180
" 01 " 90
" 00 " 0
0 ) 2 cos(
) (
f or
f or
f or
f or
T t t f A
t s
i
b i c
i

|
| t
It is highly unlikely for the channel noise to cause the phase to
shift more than 90 degree.
So if the channel noise cause a phase transition of 90 degree, it
will incur only one bit error.
Therefore the bit error rate (BER) is similar to BPSK scheme.
Hence P
e
for QPSK is similar to BPSK, as shown below:
}
Dibit coded using
Gray code
|
|
.
|

\
|
=
2
T V
erfc
2
1
P
b
2
e
Chapter 6: Digital Modulation 37
Comparison of QPSK with BPSK
As mentioned earlier, the symbol rate for QPSK is half that of
the bit rate of BPSK.

A PSK signal bandwidth depends on the rate of phase change,
the higher the phase change rate the larger the signal
bandwidth.

Since QPSK phase change rate is half that of BPSK, it occupies
half the bandwidth of BPSK signal.

If Gray code is used in coding QPSK dibits, the BER of QPSK is
similar to that of BPSK.

Important goals of digital communication systems is to make
signal with as small a bandwidth as possible so that more such
signals can go through the same transmission channel through
multiplexing. The signal modulation scheme should also allow
low BER.

Hence QPSK is preferred over BPSK in many applications.
Chapter 6: Digital Modulation 38
Mary PSK (MPSK)
In MPSK, the phase of the carrier takes one of M possible values,
namely, i = 360i/M degree, where i = 0, 1, 2, , M-1 .
Accordingly, during each symbol interval of duration T
s
, one of M
possible symbols:
1 , ... , 2 , 1 , 0
360
2 cos ) ( =
|
|
.
|

\
|
+ = M i
M
i
t f A t s
c i

t
For example for QPSK, M = 4 so that
( ) 3 , 2 , 1 , 0 90 2 cos ) ( = + = i i t f A t s
c i

t
For large M, although the transmission signal bandwidth is
small, P
e
is larger unless the signal increases in power.
More complex equipment is needed for larger M.
Chapter 6: Digital Modulation 39
Quadrature Amplitude Modulation (QAM)
This scheme is a combination of ASK and PSK, resulting in more
efficient (smaller) signal bandwidth and better P
e
if compared
with Mary PSK (M > 4).
A constellation diagram for 16-QAM is shown below.
Each symbol (a point in the diagram) uses 4 bits.
The in-phase axis (x-axis) and the quadrature-phase axis (y-
axis) each has 4 amplitude levels resulting in a combination of
16 symbols (indicated by 16 points in the diagram).
Chapter 6: Digital Modulation 40
Comparison of Digital Modulation Systems
Choice of digital modulation methods is dependent mainly on error
performance, bandwidth efficiency (in bps/Hz) and equipment
complexity.
Chapter 6: Digital Modulation 41
Chapter 6: Digital Modulation 42
Chapter 6: Digital Modulation 43
Applications
Cable Data Modems
Digital Radio
Digital Communications by Satellite
Refer to your Bound Notes.

You might also like