This action might not be possible to undo. Are you sure you want to continue?
Chapter 2 ContinuousWave
Modulation
“Analog Modulation” is the subject concerned
in this chapter.
© PoNing Chen@cm.nctu Chapter 22
2.1 Introduction
Analog communication system
The most common carrier is the sinusoidal wave.
Carrier wave
(Analog)
2
© PoNing Chen@cm.nctu Chapter 23
2.1 Introduction
Modulation
A process by which some characteristic of a carrier
is varied in accordance with a modulating wave
(baseband signal).
Sinusoidal ContinuousWave (CW) modulation
Amplitude modulation
Angle modulation
© PoNing Chen@cm.nctu Chapter 24
2.1 Introduction
Amplitude Modulation
Frequency Modulation
Baseband signal
Sinusoidal carrier
3
© PoNing Chen@cm.nctu Chapter 25
2.2 DoubleSideband with Carrier or simply
Amplitude Modulation
Two required conditions on amplitude sensitivity
1 + k
a
m(t) ≥ 0, which is ensured by k
a
m(t) ≦ 1.
The case of k
a
m(t) > 1 is called overmodulation.
The value of k
a
m(t) is sometimes represented by “percentage”
(because it is limited by 1), and is named (k
a
m(t)×100)%
modulation.
f
c
>> W, where W is the message bandwidth.
Violation of this condition will cause nonvisualized envelope.
index modulation or y sensitivit amplitude is where
), 2 cos( )] ( 1 [ ) ( Signal Modulated
) ( Baseband
) 2 cos( ) ( Carrier
a
c a c
c c
k
t f t m k A t s
t m
f A t c
π
π
+ =
=
© PoNing Chen@cm.nctu Chapter 26
2.2 Overmodulation
overmodulaion
1  ) (  ≤ t m k
a
4
© PoNing Chen@cm.nctu Chapter 27
2.2 Example of NonVisualized Envelope
© PoNing Chen@cm.nctu Chapter 28
2.2 Example of Visualized Envelope
5
© PoNing Chen@cm.nctu Chapter 29
2.2 Transmission Bandwidth
    ) ( ) (
2
) ( ) (
2
) (
) 2 cos( )] ( 1 [ ) (
c c
c a
c c
c
c a c
f f M f f M
A k
f f f f
A
f S
t f t m k A t s
+ + − + + + − = ⇒
+ =
δ δ
π
2W
Transmission bandwidth B
T
= 2W.
© PoNing Chen@cm.nctu Chapter 210
2.2 Transmission Bandwidth
Transmission bandwidth of an AM wave
For positive frequencies, the highest frequency
component of the AM wave equals f
c
+ W, and the
lowest frequency component equals f
c
– W.
The difference between these two frequencies
defines the transmission bandwidth B
T
for an AM
wave.
6
© PoNing Chen@cm.nctu Chapter 211
2.2 Transmission Bandwidth
The condition of f
c
> W ensures that the sidebands
do not overlap.
© PoNing Chen@cm.nctu Chapter 212
2.2 Negative Frequency
Operational meaning of “negative frequency”
in spectrum
If timedomain signal is realvalued, the negative
frequency spectrum is simply a mirror of the
positive frequency spectrum.
We may then define a onesided spectrum as
Hence, if only realvalued signal is considered, it is
unnecessary to introduce “negative frequency”.
. 0 for ) ( 2 ) (
sided  one
≥ = f f S f S
7
© PoNing Chen@cm.nctu Chapter 213
2.2 Negative Frequency
So the introduction of negative frequency part is
due to the need of imaginary signal part.
Signal phase information is embedded in imaginary
signal part of the signal.
phase
m
I
(t)
m
Q
(t)
© PoNing Chen@cm.nctu Chapter 214
2.2 Negative Frequency
As a result, the following two spectrums contain the
same frequency components but different phases
(90 degree shift in complex plane).
f
c
f
c
− f
c
− f
c
) (
c
f f − δ ) (
c
f f + δ
) (
c
f f − δ
) (
c
f f + −δ
) 2 cos( 2 t f
c
π
) 2 sin( 2 t f j
c
π
8
© PoNing Chen@cm.nctu Chapter 215
2.2 Negative Frequency
Summary
Complexvalued baseband signal consists of
information of amplitude and phase; while real
valued baseband signal only contains amplitude
information.
Onesided spectrum only bears amplitude
information, while twosided spectrum (with
negative frequency part) carries also phase
information.
© PoNing Chen@cm.nctu Chapter 216
2.2 Virtues of Amplitude Modulation
AM receiver can be implemented in terms of simple
circuit with inexpensive electrical components.
E.g., AM receiver
)] 4 cos( 1 [ )] ( 1 [
2
) 2 ( cos )] ( 1 [ ) ( ) (
2
2
2 2 2 2
1
t f t m k
A
t f t m k A t s t v
c a
c
c a c
π
π
+ + =
+ = =
9
© PoNing Chen@cm.nctu Chapter 217
2.2 Virtues of Amplitude Modulation
The bandwidth of m
2
(t) is twice of m(t). (So to
speak, the bandwidth of m(f)*m(f) is twice of m(f).)
Lowpass filter
© PoNing Chen@cm.nctu Chapter 218
2.2 Virtues of Amplitude Modulation
So if 2f
c
> 4W,
) (
2
mean zero is ) ( if
)] ( 1 [
2
) (
)] ( 1 [
2
) (
block DC
3
2
2
2
t m
k A
t m
t m k
A
t v
t m k
A
t v
a c
a
c
a
c
⇒
+ = ⇒
+ = ⇒
By means of a squarer, the receiver can recover the
informationbearing signal without the need of a local carrier.
10
© PoNing Chen@cm.nctu Chapter 219
2.2 Limitations of Amplitude Modulation
(DSBC)
Wasteful of power and bandwidth
Waste of power in the informationless “withcarrier” part.
) 2 cos( ) ( ) 2 cos(
) 2 cos( )] ( 1 [ ) (
carrier with
t f t m k t f A
t f t m k A t s
c a c c
c a c
π π
π
+ =
+ =
4 43 4 42 1
© PoNing Chen@cm.nctu Chapter 220
2.2 Limitations of Amplitude Modulation
Wasteful of power and bandwidth
Only requires half of bandwidth after modulation
11
© PoNing Chen@cm.nctu Chapter 221
2.3 Linear Modulation
Definition
Both s
I
(t) and s
Q
(t) in
s(t) = s
I
(t)cos(2πf
c
t) – s
Q
(t) sin(2πf
c
t)
are linear function of m(t).
© PoNing Chen@cm.nctu Chapter 222
2.3 Linear Modulation
For a single realvalued m(t), three types of
modulations can be identified according to
how s
Q
(t) are linearly related to m(t), at the
case that s
I
(t) is exactly m(t):
(Some modulation may have m
I
(t) and m
Q
(t) that
respectively bear independent information.)
1. Double SideBandSuppressed Carrier modulation
(DSBSC)
2. Single SideBand (SSB) modulation
3. Vestigial SideBand (VSB) modulation
12
© PoNing Chen@cm.nctu Chapter 223
2.3 DSBSC and SSB
Lower side band transmission m(t) SSB
Upper side band transmission m(t) SSB
0 m(t) DSBSC
s
Q
(t) s
I
(t) Type of modulation
) ( ˆ t m
) ( ˆ t m −
DSBSC
SSB
SSB
) ( of ansform Hilbert tr ) ( ˆ * t m t m =
usb
lsb
, which is used to completely “suppress” the other sideband.
© PoNing Chen@cm.nctu Chapter 224
2.3 DSBSC
Different from DSBC, DSBSC s(t) undergoes
a phase reversal whenever m(t) crosses zero.
) 2 cos( ) ( ) ( t f t m t s
c
π =
Require a receiver that can
recognize the phase reversal!
13
© PoNing Chen@cm.nctu Chapter 225
2.3 Coherent Detection for DSBSC
For DSBSC, we can no longer use the “envelope
detector” (as used for DSBC), in which no local
carrier is required for the receiver.
The coherent
detection or
synchronous
demodulation
becomes necessary.
© PoNing Chen@cm.nctu Chapter 226
2.3 Coherent Detection for DSBSC
. provided
), ( ) cos(
2
1
'
W f
t m A A
c
c c
LowPass
>
→ φ
) ( ) cos(
2
1
) ( ) 4 cos(
2
1
) ( ) 2 cos( ) 2 cos(
) ( ) 2 cos( ) (
' '
'
'
t m A A t m t f A A
t m t f t f A A
t s t f A t v
c c c c c
c c c c
c c
φ φ π
φ π π
φ π
+ + =
+ =
+ =
14
© PoNing Chen@cm.nctu Chapter 227
2.3 Coherent Detection for DSBSC
Quadrature null effect of the coherent detector.
If φ = π/2 or –π/2, the output of coherent detector for DSB
SC is nullified.
If φ is not equal to either π/2 or –π/2, the output of coherent
detector for DSBSC is simply attenuated by a factor of
cos(φ), if φ is a constant, independent of time.
However, in practice, φ often varies with time; therefore, it is
necessary to have an additional mechanism to maintain the
local carrier in the receiver in perfect synchronization with
the local carrier in the transmitter.
Such an additional mechanism adds the system complexity of
the receiver.
© PoNing Chen@cm.nctu Chapter 228
2.3 Costas Receiver for DSBSC
An exemplified
design of
synchronization
mechanism is the
Costas receiver,
where two
coherent
detectors are
used.
Inphase coherent detector
Quadraturephase coherent detector
15
© PoNing Chen@cm.nctu Chapter 229
2.3 Costas Receiver for DSBSC
Conceptually, the Costas receiver adjusts the
phase φ so that it is close to 0.
When φ drifts away from 0, the Qchannel output
will have the same polarity as the Ichannel output
for one direction of phase drift, and opposite
polarity for another direction of phase drift.
The phase discriminator then adjusts φ through the
voltage controlled oscillator.
pi/2 0 pi/2
) sin(φ
) cos(φ
© PoNing Chen@cm.nctu Chapter 230
2.3 SingleSideband Modulation
How to generate SSB signal?
1. Product modulator to generate DSBSC signal
2. Bandpass filter to pass only one of the sideband
and suppress the other.
The above technique may not be applicable to a
DSBSC signal like below. Why?
SSB
filter filter
16
© PoNing Chen@cm.nctu Chapter 231
2.3 SingleSideband Modulation
For the generation of an SSB modulated signal
to be possible, the message spectrum must have
an energy gap centered at the origin.
© PoNing Chen@cm.nctu Chapter 232
2.3 SingleSideband Modulation
Example of signal with −300 Hz ~ 300 Hz
energy gap
Voice : A band of 300 to 3100 Hz gives good
articulation
Also required for SSB modulation is a highly
selective filter
17
© PoNing Chen@cm.nctu Chapter 233
2.3 SingleSideband Modulation
Phase synchronization is also an important
issue for SSB demodulation. This can be
achieved by:
Either a separate lowpower pilot carrier
Or a highly stable local oscillator (for voice
transmission)
Phase distortion that gives rise to a Donald Duck voice
effect is relatively insensitive to human ear.
© PoNing Chen@cm.nctu Chapter 234
2.3 Vestigial Sideband Modulation
Instead of transmitting only one sideband as
SSB, VSB modulation transmits a partially
suppressed sideband and a vestige of the other
sideband.
18
SSB
VSB
© PoNing Chen@cm.nctu Chapter 235
Still, no information loss in principle.
© PoNing Chen@cm.nctu Chapter 236
2.3 Requirements for VSB filter
1. The sum of values of the magnitude response H(f) at any
two frequencies equally displaced above and below f
c
is
unity. I.e., H(f
c
− f) + H(f
c
+ f) = 1 for −f
v
< f < f
v
.
2. H(f − f
c
) + H(f + f
c
) = 1 for −W< f < W.
f
c
−f
c
So the transmission band of VSB filter is B
T
= W+ f
v
.
f
f+f
c
f−f
c
19
© PoNing Chen@cm.nctu Chapter 237
2.3 Generation of VSB Signal
Analysis of VSB
Give a real baseband signal m(t) of bandwidth W.
Then,
Let where
.   for 0 ) ( and ) ( ) (
*
W f f M f M f M > = = −
, 2 / ] / ) ( 1 )[ ( ) ( j f H f M f M
Q VSB
+ =
). ( ) ( and ,
0 , 0
0 ), 1 , 0 (
, 1
) (
1
*
f H f H
f
f f
f f
f H
j
Q Q v
v
Q
= −
¦
¹
¦
´
¦
=
< < − ∈
− ≤
=
The filter is denoted by H
Q
is because it is used to generate s
Q
(t) (cf. slide Chapter 223)
© PoNing Chen@cm.nctu Chapter 238
2.3 Generation of VSB Signal
1.0
−W
) (
1
) ( f H
j
f L
Q Q
=
). (
) ( ) (
1
) (
1
) (
1
) ( real. is ) (
1
) (
*
*
*
f L
f L f H
j
f H
j
f H
j
f L f H
j
f L
Q
Q Q Q Q Q Q Q
− =
− =
(
¸
(
¸
− = = − = − ⇒ =
2 / ) (
1
1 2 / )] ( 1 [


.

\

+ = + f H
j
f L
Q Q
20
© PoNing Chen@cm.nctu Chapter 239
2.3 How to recover from VSB signal?
( )
( )
( )
). ( ) ( because ), (
)] ( ) ( 2 )[ (
2
1
) ( ) ( and real, is )] ( 1 [ since
, )] ( 1 )[ ( )] ( 1 )[ (
2
1
)] ( 1 )[ ( )] ( 1 )[ (
2
1
) ( ) (
*
*
f L f L f M
f L f L f M
f M f M f L
f L f M f L f M
f L f M f L f M
f M f M
Q Q
Q Q
Q
Q Q
Q Q
VSB VSB
− = − =
− + + =
= − − +
− + + + =
− + − + + =
− +
∗
∗
© PoNing Chen@cm.nctu Chapter 240
2.3 VSB upper sideband transmission


.

− ≥
+ − +
× + ≤ ≤ − + − +
+ − ≤
\

+ +
× + − ≤ ≤ − − + =


.

\

+ − +
+ − +
+ +
+ =
− + + =
+ − + + = − + + =
} {
2
)] ( 1 [
} { ) (
} {
2
)] ( 1 [
} { ) (
2
1
2
)] ( 1 [
) (
2
)] ( 1 [
) (
2
1
2 / )] ( ) ( [ ) (
2 / )] ( ) ( [ 2 / )] ( ) ( [ ) (
*
*
*
*
W f f
f f L
W f f W f f f M
W f f
f f L
W f f W f f f M
f f L
f f M
f f L
f f M
f f M f f M f s
f f M f f M f f M f f M f s
c
c Q
c c c
c
c Q
c c c
c Q
c
c Q
c
c VSB c VSB VSB
c c c c DSB
1 1
1 1
) (t m
VSB
) (t s
VSB
21
( )
( )
( )
( ) ) ( ) ( 2
2
1
) (
} { )] ( 1 [ } { )] ( 1 [
2
1
) (
} {
2
)] ( 1 [
} {
2
)] ( 1 [
) (
} {
2
)] ( 1 [
} {
2
)] ( 1 [
) ( ) (
2
1
} {
2
)] ( 1 [
} {
2
)] ( 1 [
} { ) ( } { ) (
2
1
*
*
cont.
c Q c Q DSB
c c Q c c Q DSB
c
c Q
c
c Q
DSB
c
c Q
c
c Q
c c
c
c Q
c
c Q
c c c c c c
f f L f f L f s
W f f f f L W f f f f L f s
W f f
f f L
W f f
f f L
f s
W f f
f f L
W f f
f f L
f f M f f M
W f f
f f L
W f f
f f L
W f f W f f f M W f f W f f f M
+ − + + + × =
≤ + − + − + + ≤ + + + × =


.

\

− ≥
+ − +
+ + − ≤
+ +
× =


.

\

− ≥
+ − +
+ + − ≤
+ +
×
+ − + + =


.

\

− ≥
+ − +
+ + − ≤
+ +
×
+ ≤ ≤ − + − + + − ≤ ≤ − − + =
1 1
1 1
1 1
1 1
1 1
(See next slide.)
© PoNing Chen@cm.nctu Chapter 241
. 0 ) ( ) ( ) ( ) ( if
)] ( ) ( )][ ( ) ( [ ) ( ) ( ) ( ) (
= =
+ + = +
f F f M f F f M
f F f F f M f M f F f M f F f M
L R R L
R L R L R R L L
} { )] ( 1 [ W f f f f L
c c Q
≤ + + + 1
} { )] ( 1 [ W f f f f L
c c Q
≤ + − + − + 1
2
2
2
} { )] ( 1 [
} { )] ( 1 [
W f f f f L
W f f f f L
c c Q
c c Q
≤ + − + − + +
≤ + + +
1
1
) (
c Q
f f L +
) (
c Q
f f L + −
1
1
2
2 ) ( ) ( + + − + +
c Q c Q
f f L f f L
Chapter 242
22
© PoNing Chen@cm.nctu Chapter 243
( )
¦
¹
¦
´
¦
=
+ + − + + =
) ( ) ( ) (
2 ) ( ) (
2
1
) ( ) (
f H f s f s
f f L f f L f s f s
DSB VSB
c Q c Q DSB VSB
( ) 2 ) ( ) (
2
1
) ( + + − + + = ⇒
c Q c Q
f f L f f L f H
Consequently,
© PoNing Chen@cm.nctu Chapter 244
1
) ( f H
1
) (
c
f f H −
−1
) (
c
f f H + −
1
) ( ) (
c c
f f H f f H + − −
23
© PoNing Chen@cm.nctu Chapter 245
1
) ( ) (
c c
f f H f f H + − −
W f f f H f f H j f H
W f f f H f f H f L
c c Q
c c Q
≤ + − − = ⇒
≤ + − − = ⇒
  for )] ( ) ( [ ) (
  for ) ( ) ( ) (
© PoNing Chen@cm.nctu Chapter 246
2.3 Mathematical Representation of VSB
signal
) ( '
2
1
) (
2
1
) ( ) (
2
1
) (
2
1
2 / )] ( 1 )[ ( ) (
f jM f M
f H f jM f M
f jH f M f M
Q
Q VSB
+ =
− =
− =
). ( ) ( ) ( ) ( ) ( ' where f L f jM f H f M f M
Q Q
− = − =
. )] ( ' [ )] ( ) ( [ ) ( ) ( ) (
) ( ) ( ) ( ) ( ) ( '
* * * * *
*
f M f L f jM f L f M j
f L f jM f L f jM f M
Q Q
Q Q
= − = − =
= − − − = −
Transform. Hilbert of extension an is This real. is ) ( ' Notably, t m
24
© PoNing Chen@cm.nctu Chapter 247
2.3 Application of VSB Modulation
Television Signals
1. The video signal exhibits a large bandwidth and
significant lowfrequency content.
Hence, no energy gap exists (SSB becomes
impractical).
VSB modulation is adopted to save bandwidth.
Notably, since a rigid control of the transmission VSB
filter at the very highpower transmitter is expensive, a
“notquite” VSB modulation is used instead (a little
waste of bandwidth to save cost).
© PoNing Chen@cm.nctu Chapter 248
2.3 Application of VSB Modulation
VSB Filter for Television Signal Transmissions
55.25 MHz 59.75 MHz
25
© PoNing Chen@cm.nctu Chapter 249
2.3 Application of VSB Modulation
As the transmission signal is not quite VSB modulated,
the receiver needs to “reshape” the received signal
before feeding it to a VSB demodulator.
© PoNing Chen@cm.nctu Chapter 250
2.3 Application of VSB Modulation
2. In order to save the cost of the receiver (in order to
use envelope detector at the receiver), an additional
carrier is added.
Notably, additional carrier does not increase
bandwidth, but just add transmission power.
( )
) 2 sin( ) ( '
2
1
) 2 cos( ) (
2
1
1
) 2 sin( ) ( ' ) 2 cos( ) (
2
1
) 2 cos( ) (
t f t m A k t f t m k A
t f t m t f t m k A t f A t s
c c a c a c
c c a c c c
π π
π π π
±
(
¸
(
¸
+ =
± + =
26
© PoNing Chen@cm.nctu Chapter 251
2.3 Application of VSB Modulation
Distortion of envelope detector



.

\

+
(
¸
(
¸
+ →
(
¸
(
¸
+ ±
+
(
¸
(
¸
+ =
43 42 1
Distortion
2 2
2
2
LowPass
2
2 2 2 2 2
2
2 2
)) ( ' (
4
1
) (
2
1
1
2
1
) 2 cos( ) 2 sin( ) (
2
1
1 ) ( '
2
1
) 2 ( sin )) ( ' (
4
1
) 2 ( cos ) (
2
1
1 ) (
t m k t m k A
t f t f t m k t m A k
t f t m A k t f t m k A t s
a a c
c c a c a
c c a c a c
π π
π π
The distortion can be compensated by reducing the amplitude
sensitivity k
a
or increasing the width of the vestigial sideband. Both
methods are used in the design of Television broadcasting system.
© PoNing Chen@cm.nctu Chapter 252
2.3 Extension Usage of DSBSC
QuadratureCarrier Multiplexing or Quadrature
Amplitude Modulation (QAM)
Synchronization is critical in QAM modulation, which is often achieved by a
separate lowpower pilot tone outside the passband of the modulated signal.
27
© PoNing Chen@cm.nctu Chapter 253
2.4 Frequency Translation
The basic operation of SSB modulation is simply a
special case of frequency translation.
So SSB modulation is sometimes referred to as frequency
changing, mixing, or heterodyning.
The mixer is a device that consists of a product modulator
followed by a bandpass filter, which is exactly what SSB
modulation does.
© PoNing Chen@cm.nctu Chapter 254
2.4 Frequency Translation
The process is named
upconversion, if f
1
+ f
is the wanted signal,
and f
1
– f is the
unwanted image
signal.
The process is named
downconversion, if f
1
− f is the wanted
signal, and f
1
+ f is
the unwanted image
signal.
28
© PoNing Chen@cm.nctu Chapter 255
2.5 FrequencyDivision Multiplexing
Multiplexing is a technique to combine a
number of independent signals into a composite
signal suitable for transmission.
Two conventional multiplexing techniques
FrequencyDivision Multiplexing (FDM)
TimeDivision Multiplexing (TDM)
Will be discussed in Chapter 3.
© PoNing Chen@cm.nctu Chapter 256
2.5 FrequencyDivision Multiplexing
29
© PoNing Chen@cm.nctu Chapter 257
Example 2.1
© PoNing Chen@cm.nctu Chapter 258
2.6 Angle Modulation
Angle modulation
The angle of the carrier is varied in accordance with
the baseband signal.
Angle modulation provides us with a practical
means of exchanging channel bandwidth for
improved noise performance.
So to speak, angle modulation can provide better
discrimination against noise and interference than
the amplitude modulation, at the expense of
increased transmission bandwidth.
30
© PoNing Chen@cm.nctu Chapter 259
2.6 Angle Modulation
Commonly used angle modulation
Phase modulation (PM)
Frequency modulation (FM)
y. sensitivit phase is where )], ( 2 cos[ ) (
p p c c
k t m k t f A t s + = π
 
 
y. sensitivit frequency is re whe
, ) ( 2 2 cos
)) ( ( 2 cos ) (
0
0
f
t
f c c
t
f c c
k
d m k t f A
d m k f A t s
∫
∫
+ =
+ =
τ τ π π
τ τ π
© PoNing Chen@cm.nctu Chapter 260
2.6 Angle Modulation
Main differences between Amplitude
Modulation and Angle Modulation
1. Zero crossing spacing of angle modulation no
longer has a perfect regularity as amplitude
modulation does.
2. Angle modulated signal has constant envelope; yet,
the envelope of amplitude modulated signal is
dependent on the message signal.
31
© PoNing Chen@cm.nctu Chapter 261
2.6 Angle Modulation
Similarity between PM and FM
PM is simply an FM with in place of m(t).
Hence, the text only discusses FM in this chapter.
∫
t
d m
0
) ( τ τ
(
¸
(
¸
+ =
∫
t
f c c FM
d m k t f A t s
0
) ( 2 2 cos ) ( τ τ π π
)] ( 2 cos[ ) ( t m k t f A t s
p c c PM
+ = π
© PoNing Chen@cm.nctu Chapter 262
2.7 Frequency Modulation
s(t) of FM modulation is a nonlinear function of m(t).
So its general analysis is hard.
To simplify the analysis, we may assume a singletone
transmission, where
) 2 cos( ) ( t f A t m
m m
π =
(
¸
(
¸
+ =
(
¸
(
¸
+ =
(
¸
(
¸
=
∫
∫ ∫
t
f c c
t
f c c
t
i c
d m k t f A
d m k f A d f A t s
0
0 0
) ( 2 2 cos
)) ( ( 2 cos ) ( 2 cos ) (
τ τ π π
τ τ π τ τ π
32
From the formula in the previous slide,
deviation. frequency the is where
) 2 cos(
) 2 cos(
) ( ) (
m f
m c
m m f c
f c i
A k f
t f f f
t f A k f
t m k f t f
= ∆
⋅ ∆ + =
+ =
+ =
π
π
(
¸
(
¸
∆
+ =
(
¸
(
¸
⋅ ∆ + =
(
¸
(
¸
=
∫
∫
) 2 sin( 2 cos
)] 2 cos( [ 2 cos
) ( 2 cos ) (
0
0
t f
f
f
t f A
d f f f A
d f A t s
m
m
c c
t
m c c
t
i c
π π
τ τ π π
τ τ π ⇒
signal. FM of
index modulation the called often is / where
m
f f ∆ = β
© PoNing Chen@cm.nctu Chapter 263
Modulation index β is the largest deviation from 2πf
c
t in
FM system.
As a result,
1. A small β corresponds to a narrowband FM.
2. A large β corresponds to a wideband FM.
© PoNing Chen@cm.nctu Chapter 264
  ) 2 sin( 2 cos ) ( t f t f A t s
m c c
π β π + =
m c c m c i c m c
f f f f t f f f t f f f f f β π β + = ∆ + ≤ ⋅ ∆ + = ≤ ∆ − = − ) 2 cos( ) (
33
© PoNing Chen@cm.nctu Chapter 265
2.7 Narrowband Frequency Modulation
 
) 2 sin( ) 2 sin( ) 2 cos(
)] 2 sin( sin[ ) 2 sin( )] 2 sin( cos[ ) 2 cos(
) 2 sin( 2 cos ) (
t f t f A t f A
t f t f A t f t f A
t f t f A t s
m c c c c
m c c m c c
m c c
π β π π
π β π π β π
π β π
− ≈
− =
+ =
(Often, β < 0.3.)
© PoNing Chen@cm.nctu Chapter 266
2.7 Narrowband Frequency Modulation
Comparison between approximate narrowband
FM modulation and AM (DSBC) modulation
) ) ( 2 cos(
2
) ) ( 2 cos(
2
) 2 cos(
) 2 cos( )] 2 cos( 1 [
) 2 cos( )] ( 1 [ ) (
t f f
A k
t f f
A k
t f A
t f t f A k A
t f t m k A t s
m
m a
m
m a
c c
c m m a c
c a c AM
− + + + =
+ =
+ =
π π π
π π
π
) ) ( 2 cos(
2
) ) ( 2 cos(
2
) 2 cos(
) 2 sin( ) 2 sin( ) 2 cos( ) (
t f f
A
t f f
A
t f A
t f t f A t f A t s
m c
c
m c
c
c c
m c c c c FM
− − + + =
− ≈
π
β
π
β
π
π β π π
34
© PoNing Chen@cm.nctu Chapter 267
2.7 Narrowband Frequency Modulation
Represent them in terms of their lowpass
isomorphism.
)] 2 sin( ) 2 [cos(
2
)] 2 sin( ) 2 [cos(
2
) 0 ( ) (
~
t f j t f
A k
t f j t f
A k
j A t s
m m
m a
m m
m a
c AM
π π
π π
− +
+ + + =
)] 2 sin( ) 2 [cos(
2
)] 2 sin( ) 2 [cos(
2
) 0 ( ) (
~
t f j t f
A
t f j t f
A
j A t s
m m
c
m m
c
c FM
π π
β
π π
β
− −
+ + + =
© PoNing Chen@cm.nctu Chapter 268
2.7 Narrowband Frequency Modulation
Phaser diagram
)] 2 sin( ) 2 [cos(
2
t f j t f
A
m m
c
π π
β
− −
) 0 ( j A
c
+
)) 2 sin(
) 2 (cos(
2
t f j
t f
A
m
m
c
π
π
β
+
) (
~
t s
FM
)] 2 sin( ) 2 [cos(
2
t f j t f
A
m m
c
π π
β
−
) (
~
t s
AM
. Let
m a c
A k A = β
35
© PoNing Chen@cm.nctu Chapter 269
2.7 Spectrum of SingleTone FM
Modulation
 
  ( ) { }
{ } ) 2 exp( ) (
~
Re
) 2 sin( 2 exp Re
) 2 sin( 2 cos ) (
t f j t s
t f t f j A
t f t f A t s
c
m c c
m c c
π
π β π
π β π
=
+ =
+ =
  ( ) ) 2 sin( exp ) (
~
t f j A t s
m c
π β = ⇒
∑
∞
−∞ =
= ⇒
n
nt f j
n c
m
e J A t s
π
β
2
) ( ) (
~
) sin(
) (
φ φ jx
n
jn
n
e e x J =
∑
∞
−∞ =
kind. first the of function el order Bess nth the is ) ( where ⋅
n
J
(Slide Chapter 1247)
© PoNing Chen@cm.nctu Chapter 270
2.7 Spectrum of SingleTone FM
Modulation
∑
∑
∫
∫
∑
∫
∞
−∞ =
∞
−∞ =
∞
∞ −
− −
∞
∞ −
−
∞
−∞ =
∞
∞ −
−
− =
=

.

\

=
=
n
m n c
n
t nf f j
n c
ft j
n
nt f j
n c
ft j
nf f J A
dt e J A
dt e e J A
dt e t s f S
m
m
) ( ) (
) (
) (
) (
~
) (
~
) ( 2
2 2
2
δ β
β
β
π
π π
π
⇒
36
© PoNing Chen@cm.nctu Chapter 271
2.7 Spectrum of SingleTone FM
Modulation
 
 
 
∑
∑
∞
−∞ =
∞
−∞ =
+ + + − − =
− − − + − − =
− − + − =
n
m c m c n
c
n
m c m c n
c
c c
nf f f nf f f J
A
nf f f nf f f J
A
f f S f f S f S
) ( ) ( ) (
2
) ( ) ( ) (
2
) (
~
) (
~
2
1
) (
*
δ δ β
δ δ β
Consequently,
© PoNing Chen@cm.nctu Chapter 272
2.7 Spectrum of SingleTone FM
Modulation
The power of s(t)
By definition, the timeaverage autocorrelation function
is given by:
Hence, the power of s(t) is equal to:
∫ ∫
− ∞ → − ∞ →
+ = + =
T
T T
T
T T
s
dt t s t s
T
dt t s t s E
T
R ) ( ) (
2
1
lim )] ( ) ( [
2
1
lim ) ( τ τ τ
 
 
2 2
) 2 sin( 2 4 cos 1
2
1
lim
) 2 sin( 2 cos
2
1
lim ) (
2
1
lim ) 0 (
2
2
2 2 2
c
T
T
m c
c
T
T
T
m c c
T
T
T T
s
A
dt
t f t f
A
T
dt t f t f A
T
dt t s
T
R
≈
+ +
=
+ = =
∫
∫ ∫
− ∞ →
− ∞ → − ∞ →
π β π
π β π
37
© PoNing Chen@cm.nctu Chapter 273
2.7 Spectrum of SingleTone FM
Modulation
The timeaverage power spectral density of a
deterministic signal s(t) is given by
From
we obtain:
) ( ) (
2
1
lim ) (
*
2
f S f S
T
f PSD
T
T ∞ →
=
{ }.   ) ( of ransform Fourier t the is ) ( where
2
T t t s f S
T
≤ ⋅ 1
 
∑
∞
−∞ =
+ + + − − =
n
m c m c n c T
nf f f T nf f f T J T A f S )) ( sinc(2 )) ( 2 ( sinc ) ( ) (
2
β
{ }
∑
∞
−∞ =
≤ ⋅ + =
n
m c n c T
T t t nf f J A t s   ) ) ( 2 cos( ) ( ) (
2
1 π β
 
 
∑
∑
∞
−∞ =
∞
−∞ =
∞ →
∞ →
+ + + − − ×
+ + + − − =
=
n
m m c m m c n
k
m c m c k
c
p
T
T
f nf f f p f nf f f p J
kf f f kf f f J
A
f S f S
T
f PSD
) / ) ( sinc( ) / ) ( ( sinc ) (
) ( ) ( ) (
4
lim
) ( ) (
2
1
lim
) (
2
*
2
β
δ δ β
For simplicity, assume that 2T increases along the multiple of
1/f
m
., i.e., 2T = p/f
m
, where p is an integer. Also assume that f
c
is a multiple of f
m
, i.e., f
c
= qf
m
., where q is an integer. Then
m c
nf f +
m c
f n f ) 1 ( + +
m c
f n f ) 1 ( − +
p
f
nf f
m
m c
+ +
© PoNing Chen@cm.nctu Chapter 274
38
)
`
¹
¹
´
¦
+ + + − − ≈
)
`
¹
+ + + + + +
¹
´
¦
− − + − − =
)
`
¹
+ + + + +
− − + + +
+ + − − +
¹
´
¦
− − − − =
∑
∑ ∑
∑ ∑
∑ ∑
∑ ∑
∑ ∑
∑ ∑
∞
−∞ =
∞
−∞ =
− −
∞
−∞ =
− −
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞
−∞ =
∞ →
)] ( ) ( [ ) (
4
) ( ) ( ) ( ) ( ) (
) ( ) ( ) ( ) ( ) (
4
) ( ) ( ) / ) ( sinc( ) (
) ( ) ( ) / ) ( sinc( ) (
) ( ) ( ) / ) ( ( sinc ) (
) ( ) ( ) / ) ( ( sinc ) ( lim
4
) (
2
2
2
2
2
2
2
2
m c m c
n
n
c
m c
n
n m c q n
n
n
m c q n
n
n m c
n
n
c
k
m c k m m c
n
n
k
m c k m m c
n
n
k
m c k m m c
n
n
k
m c k m m c
n
n
p
c
nf f f nf f f J
A
nf f f J nf f f J J
nf f f J J nf f f J
A
kf f f J f nf f f p J
kf f f J f nf f f p J
kf f f J f nf f f p J
kf f f J f nf f f p J
A
f PSD
δ δ β
δ β δ β β
δ β β δ β
δ β β
δ β β
δ β β
δ β β
© PoNing Chen@cm.nctu Chapter 275
© PoNing Chen@cm.nctu Chapter 276
2.7 Average Power of SingleTone FM
Signal
Hence, the power of a singletone FM signal is
given by:
Question: Can we use 2∆f to be the bandwidth of a
singletone FM signal?
2
) ( ) ( ) 1 ( 1
2
) ( ) ( ) (
2
) (
2
2
2
2
2
2
c
n
q n n
n
c
n
q n n
n
n
c
A
J J
A
J J J
A
df f PSD
≈

.

\

− + =

.

\

+ =
∑
∑ ∑
∫
∞
−∞ =
+
∞
−∞ =
− −
∞
−∞ =
∞
∞ −
β β
β β β
39
© PoNing Chen@cm.nctu Chapter 277
Example 2.2
Fix f
m
and k
f
,
but vary β = ∆f/f
m
=
k
f
A
m
/f
m
.
© PoNing Chen@cm.nctu Chapter 278
Example 2.2
Fix A
m
and k
f
,
but vary β = ∆f/f
m
= k
f
A
m
/f
m
.
40
© PoNing Chen@cm.nctu Chapter 279
2.7 Spectrum of Narrowband SingleTone
FM Modulation
.
2 for 0 ) (
2
) (
1 ) (
small, is When
1
0
¦
¦
¹
¦
¦
´
¦
> ≈
≈
≈
n J
J
J
n
β
β
β
β
β
© PoNing Chen@cm.nctu Chapter 280
2.7 Spectrum of Narrowband SingleTone
FM Modulation
This results in an approximate spectrum for
narrowband singletone FM signal spectrum as
 
 
 
  ) ( ) ( ) (
4
) ( ) ( ) (
4
) ( ) ( ) (
4
) ( ) ( ) (
4
) (
2
1
2
2
0
2
2
1
2
2
2
m c m c
c
c c
c
m c m c
c
n
m c m c n
c
f f f f f f J
A
f f f f J
A
f f f f f f J
A
nf f f nf f f J
A
f PSD
+ + + − − +
+ + − +
− + + + − ≈
+ + + − − ≈
−
∞
−∞ =
∑
δ δ β
δ δ β
δ δ β
δ δ β
41
 
 
 
 
 
  ) ( ) (
16
) ( ) (
4
) ( ) (
16
) ( ) ( ) (
4
) ( ) ( ) (
4
) ( ) ( ) (
4
) (
2 2
2
2 2
2
1
2
2
0
2
2
1
2
m c m c
c
c c
c
m c m c
c
m c m c
c
c c
c
m c m c
c
f f f f f f
A
f f f f
A
f f f f f f
A
f f f f f f J
A
f f f f J
A
f f f f f f J
A
f PSD
+ + + − − +
+ + − +
− + + + − =
+ + + − − +
+ + − +
− + + + − =
δ δ
β
δ δ
δ δ
β
δ δ β
δ δ β
δ δ β
© PoNing Chen@cm.nctu Chapter 281
) ( ) ( ) ( ) 1 ( ) (
2 2
β β β β
n n n
n
n
J J J J
− −
= ⇒ − =
© PoNing Chen@cm.nctu Chapter 282
4
2
c
A
16
2 2
c
A β
4
2
c
A
16
2 2
c
A β
c
f
c
f −
m
f
16
2 2
c
A β
16
2 2
c
A β
42
© PoNing Chen@cm.nctu Chapter 283
2.7 Transmission Bandwidth of FM signals
Carson’s rule – An empirical bandwidth
An empirical rule for Transmission Bandwidth of
FM signals
For large β, the bandwidth is essentially 2∆f.
For small β, the bandwidth is effectively 2f
m
.
So Carson proposes that:


.

\

+ ∆ = + ∆ ≈
β
1
1 2 2 2 f f f B
m T
© PoNing Chen@cm.nctu Chapter 284
2.7 Transmission Bandwidth of FM signals
“Universalcurve” transmission bandwidth
The transmission bandwidth of an FM wave is the
minimum separation between two frequencies
beyond which none of the side frequencies is
greater than 1% of the carrier amplitude obtained
when the modulation is removed.
 
∑
∞
−∞ =
+ + + − − =
n
m c m c n
c
nf f f nf f f J
A
f S ) ( ) ( ) (
2
) ( δ δ β
  ) ( ) (
2
) 2 cos(
c c
c
c c
f f f f
A
t f A + + − → δ δ π
43
.
2
01 . 0  ) ( 
2
: where , 2
max max
)
`
¹
¹
´
¦
> = = ⇒
c
n
c
m T
A
J
A
n n f n B β
0.1
0.3
0.5
1.0
2.0
5.0
10.0
20.0
30.0
2
4
4
6
8
16
28
50
70
β
max
2n
β β
max max
2 2 n
f
f n
f
B
m
m T
= =
∆
© PoNing Chen@cm.nctu Chapter 285
. larger a causes smaller a , fixed For
T
B f β ∆ ⇒
20.0
13.3
8.0
6.0
4.0
3.2
2.8
2.5
2.3
f B
T
∆ /
© PoNing Chen@cm.nctu Chapter 286
2.7 Bandwidth of a General FM wave
Now suppose m(t) is no longer a single tone but a
general message signal of bandwidth W.
Hence, the “worstcase” tone is f
m
= W, and D = ∆f / W.
For nonsinusoidal modulation, the deviation ratio D is
used instead of the modulation index β.
The derivation ratio D plays the same role for
nonsinusoidal modulation that the modulation index β
plays for the case of sinusoidal modulation.
We can then use Carson’s rule or “universal curve” to
determine the transmission bandwidth B
T
.
44
© PoNing Chen@cm.nctu Chapter 287
2.7 Bandwidth of a General FM wave
Final notes
Carson’s rule usually underestimates the
transmission bandwidth.
Universal curve is too conservative in bandwidth
estimation.
So, a choice of a transmission bandwidth in
between is acceptable for most practical purposes.
© PoNing Chen@cm.nctu Chapter 288
Example 2.3
FM radio in North America requires the
maximum frequency derivation ∆f = 75 kHz.
If some message signal has bandwidth W= 15 kHz,
then the deviation ratio D = 75/15 = 5.
Then
kHz 180
1
1 75 2
1
1 2
Carson ,
=

.

\

+ × =

.

\

+ ∆ =
D D
f B
T
kHz 240 75
5
16 2
max
Curve Universal ,
= = ∆ = f
D
n
B
T
45
© PoNing Chen@cm.nctu Chapter 289
Example 2.3
In practice, a bandwidth of 200 kHz is allocated to
each FM transmitter.
So Carson’s rule underestimates B
T
, while
“Universal Curve” overestimates B
T
.
© PoNing Chen@cm.nctu Chapter 290
2.7 Generation of FM Signals
Direct FM
Carrier frequency is directly varied in accordance with the
message signal as accomplished using a voltagecontrolled
oscillator.
Indirect FM
The message is first integrated and sent to a phase modulator.
So, the carrier frequency is not directly varied in accordance to
the message signal.
46
© PoNing Chen@cm.nctu Chapter 291
2.7 Generation of FM Signals
∫
τ τ d m ) (
 
∫
+ = τ τ π π d m k t f A t s
f c c
) ( 2 2 cos ) (
(See next slide.)
© PoNing Chen@cm.nctu Chapter 292
2.7 Generation of FM Signals
Frequency multiplier
(
¸
(
¸
+ =
(
¸
(
¸
+ →
(
¸
(
¸
+ =
+ + + =
∫
∫
∑
∫
=
t
f c c
t
f c n
n
i
t
f c j
n
n
d m k t f A
d m nk t nf A
d m ik t if A
t s a t s a t s a t v
0
' ' '
0
BandPass
1
0
2
2 1
) ( 2 2 cos
) ( 2 2 cos
) ( 2 2 cos
) ( ) ( ) ( ) (
τ τ π π
τ τ π π
τ τ π π
L
(See next slide.)
47
¦
¦
¦
¦
¦
¦
¹
¦
¦
¦
¦
¦
¦
´
¦
+ + =
+ + =
+ =
+ =
L
)) cos( 10 ) 3 cos( 5 ) 5 (cos(
16
1
) ( cos
)) cos( 3 ) 2 cos( 4 ) 4 (cos(
8
1
) ( cos
)) cos( 3 ) 3 (cos(
4
1
) ( cos
) 1 ) 2 (cos(
2
1
) ( cos
5
4
3
2
x x x x
x x x x
x x x
x x
© PoNing Chen@cm.nctu Chapter 293
© PoNing Chen@cm.nctu Chapter 294
2.7 Demodulation of FM Signals
Indirect Demodulation – Phaselocked loop
(Will be introduced in Section 2.14)
Direct Demodulation
Balanced frequency discriminator
) (t s
) (
1
t s
) (
2
t s
 ) (
~

1
t s
 ) (
~

2
t s
) (
~
t s
o
differentiation filters
48
¦
¦
¦
¹
¦
¦
¦
´
¦
≤ +

.

\

− +
≤ −

.

\

+ −
=
elsewhere , 0
2
  ,
2
2
2
  ,
2
2
) (
1
T
c
T
c
T
c
T
c
B
f f
B
f f a j
B
f f
B
f f a j
f H π
π
¦
¦
¦
¹
¦
¦
¦
´
¦
≤ −

.

\

− − −
≤ +

.

\

+ + −
=
elsewhere , 0
2
  ,
2
2
2
  ,
2
2
) (
2
T
c
T
c
T
c
T
c
B
f f
B
f f a j
B
f f
B
f f a j
f H π
π
© PoNing Chen@cm.nctu Chapter 295
© PoNing Chen@cm.nctu Chapter 296
49
© PoNing Chen@cm.nctu Chapter 297
2.7 Analysis of Direct Demodulation in
terms of LowPass Equivalences
¦
¹
¦
´
¦
≤

.

\

+
=
¦
¹
¦
´
¦
≤ +
=
otherwise 0,
2
  ,
2
4
otherwise 0,
2
  ), ( 2
) (
~
1
1
T T
T
c
B
f
B
f a j
B
f f f H
f H
π
¦
¹
¦
´
¦
≤

.

\

+
= = ⇒
elsewhere , 0
2
  ), (
~
2
2
) (
~
) (
~
2
1
) (
~
1 1
T T
B
f f S
B
f a j
f S f H f S
π
(
¸
(
¸
+ = ⇒ ) (
~
) (
~
) (
~
1
t s B j
dt
t s d
a t s
T
π
(
¸
(
¸
+ =
∫
t
f c c
d m k t f A t s
0
) ( 2 2 cos ) ( τ τ π π
© PoNing Chen@cm.nctu Chapter 298
(
¸
(
¸
= ⇒
∫
t
f c
d m k j A t s
0
) ( 2 exp ) (
~
τ τ π
(
¸
(
¸
(
¸
(
¸
+ =
(
¸
(

.

\

(
¸
(
¸
+
¸

.

\

(
¸
(
¸
=
(
¸
(
¸
+ =
∫
∫
∫
t
f
T
f
c T
t
f c T
t
f f c
T
d m k j t m
B
k
aA B j
d m k j A B j
d m k j t m k jA a
t s B j
dt
t s d
a t s
0
0
0
1
) ( 2 exp ) (
2
1
) ( 2 exp
) ( 2 exp ) ( 2
) (
~
) (
~
) (
~
τ τ π π
τ τ π π
τ τ π π
π ⇒
50
© PoNing Chen@cm.nctu Chapter 299
{ }

.

\

+ +
(
¸
(
¸
+ =

.

\

+
(
¸
(
¸
+ − =
)
`
¹
¹
´
¦
(
¸
(
¸
(
¸
(
¸
+ =
=
∫
∫
∫
t
f c
T
f
c T
t
f c
T
f
c T
c
t
f
T
f
c T
c
d m k t f t m
B
k
aA B
d m k t f t m
B
k
aA B
t f j d m k j t m
B
k
aA B j
t f j t s t s
0
0
0
1 1
2
) ( 2 2 cos ) (
2
1
) ( 2 2 sin ) (
2
1
) 2 exp( ) ( 2 exp ) (
2
1 Re
) 2 exp( ) (
~
Re ) (
π
τ τ π π π
τ τ π π π
π τ τ π π
π ⇒
message. equivalent lowpass the of amplitude the obtain to
used be can detector envelope then , and 1 ) (
2
If W f t m
B
k
c
T
f
>> < ⇒
(
¸
(
¸
+ = ) (
2
1  ) (
~

1
t m
B
k
aA B t s
T
f
c T
π
¦
¹
¦
´
¦
≤

.

\

− −
=
otherwise 0,
2
  ,
2
4
) (
~
2
T T
B
f
B
f a j
f H
π
¦
¹
¦
´
¦
≤

.

\

− −
= = ⇒
elsewhere , 0
2
  ), (
~
2
2
) (
~
) (
~
2
1
) (
~
2 2
T T
B
f f S
B
f a j
f S f H f S
π
2
Similarly,
© PoNing Chen@cm.nctu Chapter 2100
(
¸
(
¸
(
¸
(
¸
− =
(
¸
(
¸
− − =
∫
t
f
T
f
c T
T
d m k j t m
B
k
aA B j
t s B j
dt
t s d
a t s
0
2
) ( 2 exp ) (
2
1
) (
~
) (
~
) (
~
τ τ π π
π ⇒
51
© PoNing Chen@cm.nctu Chapter 2101 © PoNing Chen@cm.nctu Chapter 2101
{ }

.

\

+ +
(
¸
(
¸
− =
=
∫
t
f c
T
f
c T
c
d m k t f t m
B
k
aA B
t f j t s t s
0
2 2
2
) ( 2 2 cos ) (
2
1
) 2 exp( ) (
~
Re ) (
π
τ τ π π π
π ⇒
(
¸
(
¸
− = ⇒ ) (
2
1  ) (
~

2
t m
B
k
aA B t s
T
f
c T
π
) ( 4  ) (
~
  ) (
~
 ) (
~
2 1
t m aA k t s t s t s
c f o
π = − = ⇒
Final Note: a is a filter parameter of the filters, which can be
used to adjust the amplitude of the resultant output.
© PoNing Chen@cm.nctu Chapter 2102
2.7 FM Stereo Multiplexing
How to do Stereo Transmission in FM radio?
Two requirements:
Backward compatible with monophonic radio receivers
Operate within the allocated FM broadcast channels
To fulfill these requirements, the baseband message
signal has to be remade.
) 2 cos( ) 4 cos( )] ( ) ( [ )] ( ) ( [ ) ( t f K t f t m t m t m t m t m
m m r l r l
π π + − + + =
ly. respective s, microphone right and
left by up picked signals message are ) ( and ) ( where t m t m
r l
52
© PoNing Chen@cm.nctu Chapter 2103
f
m
= 19 kHz
4 43 4 42 1 4 4 3 4 4 2 1
detection coherent For reception monophonic For
) 2 cos( ) 4 cos( )] ( ) ( [ )] ( ) ( [ ) ( t f K t f t m t m t m t m t m
m m r l r l
π π + − + + =
m
© PoNing Chen@cm.nctu Chapter 2104
m
m
Demultiplexer in receiver of FM stereo.
53
© PoNing Chen@cm.nctu Chapter 2105
2.8 Nonlinear Effects in FM Systems
The channel (including background noise, interference and
circuit imperfection) may introduce nonlinear effects on
the transmission signals.
For example, nonlinearity due to amplifiers.
¦
¦
¦
¹
¦
¦
¦
´
¦
< −
>
≤ −
=
1 ) ( ,
3
2
1 ) ( ,
3
2
1  ) (  ), (
3
1
) (
) (
3
t v
t v
t v t v t v
t v
i
i
i i i
o
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.8
1.5 1 0.5 0 0.5 1 1.5
© PoNing Chen@cm.nctu Chapter 2106
2.8 Nonlinear Effects in FM Systems
Suppose
 
¦
¦
¹
¦
¦
´
¦
=
+ =
+ + =
∫
t
f
c c i
i i i o
d m k t
t t f A t v
t v a t v a t v a t v
0
3
3
2
2 1
) ( 2 ) (
) ( 2 cos ) (
) ( ) ( ) ( ) (
τ τ π φ
φ π
Then
)] ( 2 [ cos
)] ( 2 [ cos )] ( 2 cos[
) ( ) ( ) ( ) (
3 3
1
2 2
1 1
3
3
2
2 1
t t f A a
t t f A a t t f A a
t v a t v a t v a t v
c c
c c c c
i i i o
φ π
φ π φ π
+ +
+ + + =
+ + =
54
( )
( )
4 4 4 4 3 4 4 4 4 2 1 4 4 4 4 3 4 4 4 4 2 1
4 4 4 4 4 4 3 4 4 4 4 4 4 2 1
W f B
c c
W f B
c c
W f B
c c c c
c c c
c c c c o
T T
T
t t f A a t t f A a
t t f A a A a A a
t t f t t f A a
t t f A a t t f A a t v
2 6
3
1
2 4
2
2
2 2
3
3 1
2
1
3
3
2
2 1
Carson , Carson ,
Carson ,
)] ( 3 6 cos[
4
1
)] ( 2 4 cos[
2
1
)] ( 2 cos[
4
3
2
1
)] ( 3 6 cos[ )] ( 2 cos[ 3
4
1
)] ( 2 4 cos[ 1
2
1
)] ( 2 cos[ ) (
+ ∆ = + ∆ =
+ ∆ =
+ + + +
+

.

\

+ + =
+ + + +
+ + + + =
φ π φ π
φ π
φ π φ π
φ π φ π
Thus, in order to recover s(t) from v
o
(t) using bandpass filter,
it requires:
2 / ) 2 2 ( 2 / ) 2 4 ( 2 W f f W f f
c c
+ ∆ + > + ∆ −
© PoNing Chen@cm.nctu Chapter 2107
© PoNing Chen@cm.nctu Chapter 2108
. 2 3 ly, equivalent or W f f
c
+ ∆ >
The filtered output is therefore:
)] ( 2 cos[
4
3
) (
3
3 1 filtered ,
t t f A a A a t v
c c c o
φ π +

.

\

+ =
Observations
Unlike AM modulation, FM modulation is not affected
by distortion produced by transmission through a channel
with amplitude nonlinearities.
So the FM modulation allows the usage of highly non
linear amplifiers and power transmitters.
55
© PoNing Chen@cm.nctu Chapter 2109
2.8 AMtoPM Conversion
Although FM modulation is insensitive to amplitude
nonlinearity, it is indeed very sensitive to phase
nonlinearity.
A common type of phase nonlinearity encountered in
microwave radio transmission is the AMtoPM conversion.
The AMtoPM conversion is owing to that the phase
characteristic of amplifiers (or repeaters) also depends on the
instantaneous amplitude of the input signal.
Notably, the nonlinear amplifiers discussed previously will
leave the phase of the input unchanged.
Often, it requires that the peak phase change for a 1dB
change in input envelope is less than 2%.
© PoNing Chen@cm.nctu Chapter 2110
2.9 Superheterodyne Receiver
A commercial radio communication system
contains not only the “transmission” but also
some other functions, such as:
Carrierfrequency tuning, to select the desired
signals
Filtering, to separate the desired signal from other
unwanted signals
Amplifying, to compensate for the loss of signal
power incurred in the course of transmission
56
© PoNing Chen@cm.nctu Chapter 2111
2.9 Superheterodyne Receiver
A superheterodyne receiver or superhet is designed to
facilitate the fulfillment of these functions, especially
the first two.
It overcomes the difficulty of having to build a tunable highly
selective and variable filter ( rather a fixed filter is applied on
IF section).
heterodyning function
© PoNing Chen@cm.nctu Chapter 2112
First detector
2.9 Superheterodyne Receiver
200 kHz 10 kHz IF bandwidth
10.7 MHz 0.455 MHz Midband frequency of IF section
88108 MHz 0.5351.605 MHz RF carrier range
FM Radio AM Radio Example Example
Second detector
57
© PoNing Chen@cm.nctu Chapter 2113
2.9 Image Interference
Fix f
IF
and f
LO
at the receiver end. What is the
f
RF
that will survive at the IF section output?
Answer:
Example. Suppose the receiver use 1.105MHz local
oscillator, and receives two RF signals respectively
centered at 0.65MHz and 1.56MHz.
 
IF LO RF
f f f ± =
MHz 455 . 0
IF
f
MHz 455 . 0
IF
f
MHz 65 . 0
RF
f
MHz 56 . 1
ce Interferen
f
MHz 105 . 1
LO
f
© PoNing Chen@cm.nctu Chapter 2114
MHz .455 0
RF LO IF
f f f − =
MHz .755 1
age Im
= + =
RF LO
f f f
MHz .455 0
ce Interferen Image
= − =
LO ce Interferen
f f f MHz 665 . 2 = +
LO ce Interferen
f f
2.9 Image Interference
Then at the output of the IF section, we respectively obtain:
58
© PoNing Chen@cm.nctu Chapter 2115
2.9 Image Interference
A cure of image interference is to employ a highly
selective stages in the RF session in order to favor the
desired signal (at f
RF
) and discriminate the undesired
signal (at f
RF
+ 2f
IF
or f
RF
– 2f
IF
).
© PoNing Chen@cm.nctu Chapter 2116
2.9 Advantage of Constant Envelope for FM
modulation
Observations
For FM modulation, any variation in amplitude is
caused by noise or interference.
For FM modulation, the information is resided on
the variations of the instantaneous frequency.
So we can use an amplitude limiter to remove the
amplitude variation, but to retain the frequency
variation, after the IF section.
59
© PoNing Chen@cm.nctu Chapter 2117
2.9 Advantage of Constant Envelope for FM
modulation
Amplitude limiter
Clipping the modulated wave at the IF section
output (almost to the zero axis) to result in a near
rectangular wave.
Pass the rectangular wave through a bandpass filter
centered at f
IF
to suppress harmonics (due to
clipping).
Then the filter output retains the frequency
variation with constant amplitude.
© PoNing Chen@cm.nctu Chapter 2118
2.10 Noise in CW Modulation Systems
To simplify the system analysis, we assume:
ideal bandpass filter (that is just wide enough to pass the
modulated signal s(t) without distortion),
ideal demodulator,
Gaussian distributed white Noise process.
So the only source of imperfection is from the noise.
60
© PoNing Chen@cm.nctu Chapter 2119
T
B
2.10 Noise in CW Modulation Systems
), ( ) ( ) ( t n t s t x + =
So after passing through the ideal bandpass filter, s(t) is
unchanged but w(t) becomes a narrowband noise n(t).
Hence,
). 2 sin( ) ( ) 2 cos( ) ( ) ( where t f t n t f t n t n
c Q c I
π π − =
© PoNing Chen@cm.nctu Chapter 2120
2.10 Noise in CW Modulation Systems
Input signaltonoise ratio (SNR
I
)
The ratio of the average power of the modulated
signal s(t) to the average power of the filtered noise
n(t).
Output signaltonoise ratio (SNR
O
)
The ratio of the average power of the demodulated
message signal to the average power of the noise,
measured at the receiver output.
61
© PoNing Chen@cm.nctu Chapter 2121
2.10 Noise in CW Modulation Systems
It is sometimes advantageous to look at the lowpass
equivalent model.
Channel signaltonoise ratio (SNR
C
)
The ratio of the average power of the modulated
signal s(t) to the average power of the channel noise
in the message bandwidth, measured at the receiver
input (as illustrated below).
© PoNing Chen@cm.nctu Chapter 2122
2.10 Noise in CW Modulation Systems
Notes
SNR
C
is nothing to do with the receiver structure, but
depends on the channel characteristic and modulation
approach.
SNR
O
is however receiverstructure dependent.
Finally, define the figure of merit for the receiver as:
C
O
SNR
SNR
= merit of figure
62
© PoNing Chen@cm.nctu Chapter 2123
2.11 Noise in Linear Receivers Using
Coherent Detection
Recall that for AM demodulation
When the carrier is suppressed, linear coherent
detection is used. (Section 2.11)
When the carrier is additionally transmitted,
nonlinear envelope detection is used. (Section 2.12)
The noise analysis of the above two cases are
respectively addressed in Sections 2.11 and
2.12.
© PoNing Chen@cm.nctu Chapter 2124
2.11 Noise in Linear Receivers Using
Coherent Detection
. to d bandlimite ) ( PSD and meam zero with stationary : ) ( W f S t m
M
) 2 cos( ) ( ) ( t f t m CA t s
c c
π =
) 2 cos( t f
c
π
63
© PoNing Chen@cm.nctu Chapter 2125
2.11 Noise in Linear Receivers Using
Coherent Detection
Average signal power
 
 
2
) 2 ( cos
2
1
lim
) ( ) 2 ( cos
2
1
lim
) ( ) 2 ( cos
2
1
lim )] ( [
2
1
lim
2 2
2 2 2
2 2 2 2
2 2 2 2 2
P A C
dt t f
T
P A C
dt t m E t f A C
T
dt t m t f A C E
T
dt t s E
T
c
T
T
c
T
c
T
T
c c
T
T
T
c c
T
T
T T
=
=
=
=
∫
∫
∫ ∫
− ∞ →
− ∞ →
− ∞ → − ∞ →
π
π
π
power. message the is ) ( )] ( [ where
2
∫
−
= =
W
W
M
df f S t m E P
© PoNing Chen@cm.nctu Chapter 2126
2.11 Noise in Linear Receivers Using
Coherent Detection
Noise power in the message bandwidth
0
0
2
) ( WN df
N
df f S
W
W
W
W
w
= =
∫ ∫
− −
64
© PoNing Chen@cm.nctu Chapter 2127
) 2 cos( t f
c
π
2.11 Noise in Linear Receivers Using
Coherent Detection
Channel SNR for DSCSC and coherent detection (observed at x(t))
Next, we calculate output SNR (observed at y(t)) under the condition
that the transmitter and the receiver are perfectly synchronized.
o
c
o
c
C
WN
P A C
WN
P A C
SNR
2
2
2 2 2 2
SC  DSB ,
= =
) 2 sin( ) 2 cos( ) ( ) 2 cos( ) (
) ( ) ( ) (
t f n t f t n t f t m CA
t n t s t x
c Q c I c c
π π π − + =
+ =
) (
2
1
) (
2
1
) 2 cos( ) 2 sin( ) ( ) 2 ( cos ) ( ) ( ) 2 ( cos
) 2 cos( )] 2 sin( ) ( ) 2 cos( ) ( ) ( ) 2 cos( [
) 2 cos( ) ( ) (
LowPass
2 2
t n t m CA
t f t f t n t f t n t m t f CA
t f t f t n t f t n t m t f CA
t f t x t v
I c
c c Q c I c c
c c Q c I c c
c
+ →
− + =
− + =
= ⇒
π π π π
π π π π
π
) (
2
1
) (
2
1
) ( t n t m CA t y
I c
+ = ⇒
 
 
0
2 2
2
2 2
2
2 2 2
SC  DSB ,
2 )] ( [ 4 / ) (
4 / ) (
WN
P A C
t n E
P A C
t n E
t m A C E
SNR
c c
I
c
O
= = = ⇒
© PoNing Chen@cm.nctu Chapter 2128
)].) ( [ )] ( [ )] ( [ (Recall
2 2 2
t n E t n E t n E
Q I
= =
65
. 1 detection coherent and SC  for DSB merit of Figure = ⇒
© PoNing Chen@cm.nctu Chapter 2129
Similar derivation on SSB and coherent detection yields the
same figure of merit.
Conclusions
Coherent detection for SSB performs the same as
coherent detection for DSBSC.
There is no SNR degradation for SSB and DSBSC
coherent receivers. The only effect of these modulation
and demodulation processes is to translate the message
signal to a different frequency band to facilitate its
transmission over a bandpass channel.
No tradeoff between noise performance and bandwidth.
This may become a problem when high quality
transceiving is required.
© PoNing Chen@cm.nctu Chapter 2130
2.12 Noise in AM Receivers Using Envelope
Detection
) 2 cos( )] ( 1 [ ) ( t f t m k A t s
c a c
π + =
 
mean.) zero ) ( (Assume ) 1 (
2
) 2 ( cos
2
1
lim )) ( 1 ( )] ( [
2
1
lim
2
2
2 2 2 2
t m P k
A
dt t f
T
t m k E A dt t s E
T
a
c
T
T
c
T
a c
T
T T
+ =
+ =
∫ ∫
− ∞ → − ∞ →
π
66
Hence, channel SNR for DSBC is equal to:
Next, we calculate output SNR (observed at y(t)) under the condition that
the transmitter and the receiver are perfectly synchronized.
0
0
2
) ( Also, WN df
N
df f S
W
W
W
W
w
= =
∫ ∫
− −
o
a c
C
WN
P k A
SNR
2
) 1 (
2 2
AM ,
+
= ⇒
) 2 sin( ) ( ) 2 cos( )] ( )) ( 1 ( [
) 2 sin( ) ( ) 2 cos( ) ( ) 2 cos( )] ( 1 [
) ( ) ( ) (
t f t n t f t n t m k A
t f t n t f t n t f t m k A
t n t s t x
c Q c I a c
c Q c I c a c
π π
π π π
− + + =
− + + =
+ =
© PoNing Chen@cm.nctu Chapter 2131
( )
)] ( ) ( [
2
1
)] ( )) ( 1 ( [
2
1
) ( )] ( )) ( 1 ( [
2
1
) ( ) (
block DC
2 2
LowPass
2
t n t m k A
t n t m k A
t n t n t m k A
t x t y
I a c
I a c
Q I a c
+ =
+ + ≈
+ + + =
=
envelop detector
x
x
) (
~
)] ( 1 [ if t n t m k A
a c
>> +
© PoNing Chen@cm.nctu Chapter 2132
(Refer to slide 136.)
67
 
 
0
2 2
2
2 2
2
2 2 2
AM ,
2 )] ( [ 2 / ) (
2 / ) (
WN
P k A
t n E
P k A
t n E
t m k A E
SNR
a c a c
I
a c
O
= = = ⇒
© PoNing Chen@cm.nctu Chapter 2133
1
1 ) 2 ( ) 1 (
) 2 (
2
2
2 2
0
2 2
AM ,
AM ,
<
+
=
+
≈ ⇒
P k
P k
WN P k A
WN P k A
SNR
SNR
a
a
o a c
a c
C
O
Conclusion
Even if the noise power is small compared to the
average carrier power at the envelope detector output,
the noise performance of a full AM receiver is
inferior to that of a DSBSC receiver due to the
wastage of transmitter power.
© PoNing Chen@cm.nctu Chapter 2134
Example 2.4 SingleTone Modulation
Assume
) 2 cos( )] 2 cos( 1 [ ) (
) 2 cos( ) (
t f t f A k A t s
t f A t m
c m m a c
m m
π π
π
+ = ⇒
=
 
(
)
) 1 (
2 4
0
2
1
) 2 ( cos ) 2 ( cos
) 2 ( cos ) 2 cos( 2 ) 2 ( cos
2
1
lim
) 2 ( cos ) 2 cos( 1
2
1
lim )] ( [
2
1
lim
2
2 2 2
2
2 2 2 2
2 2 2
2
2
2 2
P k
A A k
A
dt t f t f A k
t f t f A k t f
T
A
dt t f t f A k
T
A dt t s E
T
a
c m a
c
c m m a
T
T
c m m a c
T
c
T
T
c m m a
T
c
T
T T
+ = 
.

\

+ + =
+
+ =
+ =
∫
∫ ∫
− ∞ →
− ∞ → − ∞ →
π π
π π π
π π
Hence,
68
.
2 / 1
2 /
1
, discussion previous as procedure similar Following
.
2
) 2 ( cos
1
lim where
2 2
2 2
2
2
AM ,
AM ,
2
0
2 2
m a
m a
a
a
C
O
m
T
m m
T
A k
A k
P k
P k
SNR
SNR
A
dt t f A
T
P
+
=
+
≈
⇒
= =
∫
∞ →
π
So even if for 100% percent modulation (k
a
A
m
= 1), the figure of merit
= 1/3. This means that an AM system with envelope detection must
transmit three times as much average power as DSBSC with coherent
detector to achieve the same quality of noise performance.
© PoNing Chen@cm.nctu Chapter 2135
© PoNing Chen@cm.nctu Chapter 2136
2.12 Threshold Effect
What if is violated (in AM modulation
with envelope detection)?
) 2 sin( ) ( ) 2 cos( )] ( )) ( 1 ( [
) 2 sin( ) ( ) 2 cos( ) ( ) 2 cos( )] ( 1 [
) ( ) ( ) (
t f t n t f t n t m k A
t f t n t f t n t f t m k A
t n t s t x
c Q c I a c
c Q c I c a c
π π
π π π
− + + =
− + + =
+ =
) (
~
)] ( 1 [ t n t m k A
a c
>> +

~
 
~
 
~
 2 
~
 2 ) ( 
~

2 2 ) ( 
~

2 2 2 2 2 2
2 2 2 2 2 2 2
n B n B n B n B n B n n B n B
n B n B n B n n B n B n n B n B
I Q I
I I I Q I I Q I
+ ≈ + + ≈ + + = + + ⇒ <<
+ ≈ + + ≈ + + + = + + ⇒ >>
69
( )
) ) (
~
 )) ( 1 ( (
2
1
) ) (
~
 )) ( 1 ( (
2
1
 ) (
~
 )) ( 1 (  ) (
~
 2 )) ( 1 (
2
1
 ) (
~
 )) ( 1 ( ) ( 2 )) ( 1 (
2
1
) ( )] ( )) ( 1 ( [
2
1
) ( ) (
2
2 2 2
2 2 2
2 2
LowPass
2
t n t m k A
t n t m k A
t n t m k A t n t m k A
t n t m k A t n t m k A
t n t n t m k A
t x t y
a c
a c
a c a c
a c I a c
Q I a c
+ + ≈
+ + =
+ + + + ≈
+ + + + =
+ + + =
=
] ) (
~
 ) ( [
2
1
block DC
t n t m k A
a c
+ =
© PoNing Chen@cm.nctu Chapter 2137
. ) (
~
)] ( 1 [ Assume t n t m k A
a c
<< +
¹
´
¦
<< + +
>> + +
=
 ) (
~
 )] ( 1 [ , ) (
~
 ) (
 ) (
~
 )] ( 1 [ ), ( ) (
) ( 2
t n t m k A t n t m k A
t n t m k A t n t m k A
t y
a c a c
a c I a c
© PoNing Chen@cm.nctu Chapter 2138
 
 
0
2 2
2 2
2 2
2
2 2 2
AM ,
4 )] ( [ )] ( [  ) (
~

) (
WN
P k A
t n E t n E
P k A
t n E
t m k A E
SNR
a c
Q I
a c a c
O
=
+
= = ⇒
2
1
) 1 ( 2 ) 2 ( ) 1 (
) 4 (
2
2
2 2
0
2 2
AM ,
AM ,
<
+
=
+
≈ ⇒
P k
P k
WN P k A
WN P k A
SNR
SNR
a
a
o a c
a c
C
O
70
© PoNing Chen@cm.nctu Chapter 2139
2.12 Threshold Effect
Threshold effect
For AM with envelope detection, there exists a
carriertonoise ratio ρ (namely, the power ratio
between unmodulated carrier A
c
cos(2πf
c
t) and the
passband noise n(t) ) below which the noise
performance of a detector deteriorates rapidly.
¦
¦
¹
¦
¦
´
¦
<< + =
>> + =
=
 ) (
~
 )] ( 1 [ if ,
4
 ) (
~
 )] ( 1 [ if , 2
2
2
0
2 2
2
0
2 2
AM ,
t n t m k A P k
WN
P k A
t n t m k A P k
WN
P k A
SNR
a c a
a c
a c a
a c
O
ρ
ρ
0
2
0
2
4 2
2 /
WN
A
WN
A
c c
= = ρ
© PoNing Chen@cm.nctu Chapter 2140
2.12 General Formula for SNR
O
in Envelope
Detection
For envelope detector, the noise is no longer additive; so the
original definition of SNR
O
(which is based on additive
noise) may not be wellapplied.
A new definition should be given:
Definition. The (general) output signaltonoise ratio
for an output y(t) due to a carrier input is defined as
)] ( [ Var
2
t y
s
SNR
o
O
=
alone. noise of presense the in
) ( to equal is ) ( and )], ( [ )] ( [ where t y t y t y E t y E s
o o o
− =
) ( ) ( t y s t y
o o
+ =
71
© PoNing Chen@cm.nctu Chapter 2141
2.12 General Formula for SNR
O
in Envelope
Detection
s
o
is named the mean output signal.
Var[y(t)] is named the mean output noise power.
Example. y(t) = A + n
I
(t), where n
I
(t) is zero mean..
¹
´
¦
= =
= − + =
)] ( [ )] ( [ Var )] ( [ Var
)] ( [ )] ( [
2
t n E t n t y
A t n E t n A E s
I I
I I o
)] ( [
2
2
t n E
A
SNR
I
O
= ⇒
This somehow shows the backward compatibility of the new
definition.
© PoNing Chen@cm.nctu Chapter 2142
2.12 General Formula for SNR
O
in Envelope
Detection
Now for an envelope detector, the output due to a carrier
input and additive Gaussian noise channel is given by:
) ( )) ( ( ) (
2 2
t n t n A t y
Q I
+ + =
pdf with d distribute Rayleigh is ) ( ) ( ) (
2 2
t n t n t y
Q I o
+ = ⇒
. 0 for
2
exp ) (
2
2
2
) (
≥


.

\

− = y
y y
y f
N N
t y
o
σ σ
pdf with d distribute Rician is ) (t y ⇒
)]. ( [ where , 0 for
2
exp ) (
2 2
2
0
2
2 2
2
) (
t n E y
Ay
I
A y y
y f
N
N N N
t y
= ≥


.

\



.

\
 +
− = σ
σ σ σ
I
0
( ) = modified Bessel function
of the first kind of zero order.
72
© PoNing Chen@cm.nctu Chapter 2143
). 4 /( ) 2 /( and / by taking
, ) (
4
exp ) exp(
) 2 (
2
exp
) ( )] ( [
0
2 2 2 2
0
0
2
2
2 / 3
0
2
0
2
2 2
2
2
0
) (
WN A A Ay u
du u I
u
u
dy
Ay
I
A y y
dy y yf t y E
N N
N
N N N
t y
= = =


.

\

− − =


.

\



.

\
 +
− =
=
∫
∫
∫
∞
∞
∞
σ ρ σ
ρ
ρ
ρ
σ
σ σ σ
( )
( )
.
2 2
3
2
exp 2
) 2 /( by taking , exp 2
2
exp ) ( )] ( [
0
1 ) 2 / 3 (
2 2
0
2
0
2
2
2
2
0
) (
π
σ σ
σ
σ σ
σ σ
N N
N
N N
N N
t y o
dz z z
y z dz z z
dy
y y
dy y yf t y E
o
=

.

\

Γ =
− =
= − =


.

\

− = =
∫
∫
∫ ∫
∞
−
∞
∞ ∞
© PoNing Chen@cm.nctu Chapter 2144
Appendix 4 Confluent Hypergeometric
Functions
The Kummer confluent hypergeometric function is a
solution of Kummer’s equation
For b ≠ 0, −1, −2, …., the Kummer confluent
hypergeometric function is equal to
1
F
1
(a;b;x).
complex , for 0 ) (
2
2
b a ay
dx
dy
x b
dx
y d
x = − − +
.
1 ) (
) 1 ( ) 1 ( ) (
where ,
! ) ( ) ( ) (
) ( ) ( ) (
) ; ; (
function tric hypergeome d Generalize
0 0 2 1
2 1
¹
´
¦
=
− + + =
⋅ =
∑
∞
=
a
k a a a a
k
x
b b b
a a a
x b a F
k
k
k k q k k
k p k k
q p
L
L
L
r
r
. / ) 0 ( ' and 1 ) 0 ( conditions boundary with b a y y = =
73
© PoNing Chen@cm.nctu Chapter 2145
A4.2 Properties of the Confluent
Hypergeometric Function
) ; ; ( ) ; ; ( ) exp( . 5
4
1
; 1 ;
2 2
) 2 / (
) ( ) exp( . 4
. as 2
2 2
) 1 (
2
exp ) ; 1 ; 2 / 1 ( . 3
. 1 ) ; 1 ; 1 ( 2.
. 0 as 1 ) ; ; ( . 1
1 1 1 1
2
1 1
0
0
2 2 1
2 0 1 1
1 1
1 1
u F u F u
b
m
F
b
m
du u I u b u
x
x
x
xI
x
I x
x
x F
x x F
x x
b
a
x b a F
m
m
− − = ⋅ −
(
¸
(
¸

.

\

Γ
= −
∞ → ≈


.

\


.

\

+

.

\

+ ×

.

\

− = − −
− = −
→ + ≈
∫
∞
−
β α β β α
π
© PoNing Chen@cm.nctu Chapter 2146
2.12 General Formula for SNR
O
in Envelope
Detection
Hence,
(
¸
(
¸

.

\

− − =
(
¸
(
¸

.

\

− =
(
¸
(
¸

.

\

Γ
− =


.

\

− − =
−
∞
∫
ρ σ
π
ρ ρ σ
π
ρ
ρ
ρ
ρ
σ
ρ
ρ
ρ
σ
; 1 ;
2
1
2
; 1 ;
2
3
) exp(
2
; 1 ;
2
3
) 4 ( 2
) 2 / 3 (
) exp(
) 2 (
) (
4
exp ) exp(
) 2 (
)] ( [
1 1
1 1
1 1
2 / 3 2 / 3
0
0
2
2
2 / 3
F
F
F
du u I
u
u t y E
N
N
N
N
By Property 4
By Property 5
74
© PoNing Chen@cm.nctu Chapter 2147
2.12 General Formula for SNR
O
in Envelope
Detection
As a result,
Similarly, we can obtain:
. 1
2
; 1 ;
2
1
2
)] ( [ )] ( [
2
2
1 1 (
¸
(
¸
−


.

\

− − = − =
N
N o o
A
F t y E t y E s
σ
σ
π
(
(
¸
(
¸
(
¸
(
¸


.

\

− − − + =
(
(
¸
(
¸
(
¸
(
¸


.

\

− − −


.

\

− − =
2
2
2
1 1
2
2
2
2
2
2
1 1
2
2
1 1
2
2
; 1 ;
2
1
4 2
1 2
2
; 1 ;
2
1
4 2
; 1 ; 1 2 )] ( Var[
N N
N
N N
N
A
F
A
A
F
A
F t y
σ
π
σ
σ
σ
π
σ
σ
By Property 2
© PoNing Chen@cm.nctu Chapter 2148
2.12 General Formula for SNR
O
in Envelope
Detection
This concludes to:
( )  
( )  
 
 
 
¦
¦
¦
¹
¦
¦
¦
´
¦
→
+ − +
− +
∞ →
− +
−
≈
=
− − − +
− − −
=
0 as ,
) 2 / 1 ( ) 1 (
4
1 ) 2 / 1 (
as ,
/ 2 ) 1 (
4
1 / 2
2
where ,
; 1 ; 2 / 1 ) 1 (
4
1 ; 1 ; 2 / 1
) (
2
2
2
2
2
2
2
1 1
2
1 1
ρ
ρ ρ
π
ρ
ρ
π ρ ρ
π
π ρ
σ
ρ
ρ ρ
π
ρ
N
O
A
F
F
SNR
75
¹
´
¦
→
∞ →
=
¦
¹
¦
´
¦
→
−
∞ →
≈
¦
¦
¹
¦
¦
´
¦
→
+ − +
∞ → − +
=
0 as , 91 . 0
as ,
0 as ,
4 16
as ,
0 as
) 2 ( ) 1 ( 16
as ,
4
) (
2
2
2
2
ρ ρ
ρ ρ
ρ
π
πρ
ρ ρ
ρ
ρ π ρ
πρ
ρ πρ
π
ρ
O
SNR
(Continue from the previous slide.)
© PoNing Chen@cm.nctu Chapter 2149
© PoNing Chen@cm.nctu Chapter 2150
2.12 General Formula for SNR
O
in Envelope
Detection
ρ P k SNR
a O
2
AM ,
=
( )  
( )  
es. approximat limiting two the and
; 1 ; 2 / 1 ) 1 (
4
1 ; 1 ; 2 / 1
of Curve
2
1 1
2
1 1
ρ ρ
π
ρ
− − − +
− − −
=
F
F
SNR
O
ρ P k SNR
a O
2
AM ,
2 =
0.1
1
10
100
000
0.1 1 10 100
76
© PoNing Chen@cm.nctu Chapter 2151
2.12 General Formula for SNR
O
in Envelope
Detection
Remarks
For large carriertonoise ratio ρ, the envelope detector
behaves like a coherent detector in the sense that the output
SNR is proportional to ρ.
For small carriertonoise ratio ρ, the (newly defined) output
signaltonoise ratio of the envelope detector degrades faster
than a linear function of ρ (decrease at a rate of ρ
2
).
From “threshold effect” and the “general formula for SNR
O
”,
we may see that the envelope detector favors a strong signal.
This is sometimes called “weak signal suppression”.
© PoNing Chen@cm.nctu Chapter 2152
2.13 Noise in FM Receivers
To simplify the system analysis, we assume:
ideal bandpass filter (that is just wide enough to pass the
modulated signal s(t) without distortion),
ideal demodulator,
Gaussian distributed white Noise process.
So the only source of imperfection is from the noise.
77
. ) ( 2 ) ( where )], ( 2 cos[ ) (
0
∫
= + =
t
f c c
d m k t t t f A t s τ τ π φ φ π
( )
( ) )] ( 2 cos[ )] ( ) ( [ sin ) ( )] ( ) ( cos[ ) (
)] ( 2 sin[ )] ( ) ( sin[ ) (
)] ( 2 cos[ )] ( ) ( cos[ ) (
)] ( ) ( ) ( 2 cos[ ) ( )] ( 2 cos[
)] ( 2 cos[ ) ( )] ( 2 cos[ ) (
2 2 2
t t f t t t r t t t r A
t t f t t t r
t t f t t t r A
t t t t f t r t t f A
t t f t r t t f A t x
c c
c
c c
c c c
c c c
θ π φ ψ φ ψ
φ π φ ψ
φ π φ ψ
φ ψ φ π φ π
ψ π φ π
+ − + − + =
+ − −
+ − + =
− + + + + =
+ + + =
)
`
¹
¹
´
¦
− +
−
+ =
−
)] ( ) ( cos[ ) (
)] ( ) ( sin[ ) (
tan ) ( ) ( where
1
t t t r A
t t t r
t t
c
φ ψ
φ ψ
φ θ
© PoNing Chen@cm.nctu Chapter 2153
)) ( sin( ) ( ) (
)) ( cos( ) ( ) (
¹
´
¦
=
=
t t r t n
t t r t n
Q
I
ψ
ψ
( )
)] ( 2 cos[
)] ( 2 cos[ )] ( ) ( [ sin ) ( )] ( ) ( cos[ ) ( ) (
Limiter
2 2
2
t t f A
t t f t t t r t t t r A t x
c
c c
θ π
θ π φ ψ φ ψ
+ ⋅ →
+ − + − + =
© PoNing Chen@cm.nctu Chapter 2154
) ( ' 2 ) (
~
Output )) ( 2 cos( ) ( Input t aA t s t t f A t s
o c
θ θ π = + =
Recall on slide 94, we have talked about the Balanced
Frequency Discriminator, whose input and output satisfy:
Next, the signal will be passed through a Discriminator.
78
    ) ( exp ) ( ' ) (
~
) (
~
) (
~
1
t j B t aAj t s B j
dt
t s d
a t s
T T
θ π θ π + =
(
¸
(
¸
+ =
    ) ( exp ) ( ' ) (
~
) (
~
) (
~
2
t j B t aAj t s B j
dt
t s d
a t s
T T
θ π θ π − − =
(
¸
(
¸
− − =
) ( ' 2  ) (
~
  ) (
~
 ) (
~
2 1
t aA t s t s t s
o
θ = − = ⇒
Specifically, with : have we )), ( exp( ) (
~
t j A t s θ =
© PoNing Chen@cm.nctu Chapter 2155
dt
t t t r A
t t t r
t d
aA t aA t v
c


.

\

)
`
¹
¹
´
¦
− +
−
+
= =
−
)] ( ) ( cos[ ) (
)] ( ) ( sin[ ) (
tan ) (
2 ) ( ' 2 ) (
tor discrimina the through passing after Thus,
1
φ ψ
φ ψ
φ
θ
© PoNing Chen@cm.nctu Chapter 2156
( )
( )
( )
(
(
¸
(


.

− +
− −
−
\

− +
−
×
¸


.

\

− +
−
+
+ =
2
2
)] ( ) ( cos[ ) (
' )] ( ) ( cos[ ) ( )] ( ) ( sin[ ) (
)] ( ) ( cos[ ) (
' )] ( ) ( sin[ ) (
)] ( ) ( cos[ ) (
)] ( ) ( sin[ ) (
1
1
) ( ' 2 ) (
t t t r A
t t t r t t t r
t t t r A
t t t r
t t t r A
t t t r
t aA t v
c
c
c
φ ψ
φ ψ φ ψ
φ ψ
φ ψ
φ ψ
φ ψ
φ
79
© PoNing Chen@cm.nctu Chapter 2157
Assumption 1: A
c
>> r(t) with high probability.
Assumption 2: 2aA = 1/(2π).
We then make the following assumption (to simplify the derivation).
( )
( )
( )
( )
) ( 2 ) ( 2
' )] ( ) ( sin[ ) (
1
) ( 2
' )] ( ) ( sin[ ) (
) ( '
' )] ( ) ( cos[ ) ( )] ( ) ( sin[ ) (
' )] ( ) ( sin[ ) (
1 ) ( ' ) ( 2
2
t n t m k
t t t r
A
t m k
A
t t t r
t
A
t t t r t t t r
A
t t t r
t t v
d f
c
f
c
c
c
π π
φ ψ π
φ ψ
φ
φ ψ φ ψ
φ ψ
φ π
+ =
− + =
−
+ ≈


.
 − −
−
\
 −
× + ≈
We then obtain the desired “additive” form.
Omit the 2nd term by Assumption 1.
© PoNing Chen@cm.nctu Chapter 2158
There is however one problem remaining, i.e., n
d
(t) depends on
the message signal φ(t)!
Assumption 3: ψ(t) − φ(t) is uniformly distributed, and is
independent of m(t) and r(t).
The above assumption is true for any deterministic φ(t)!
It has been shown that Assumption 3 is justifiable for high
carriertonoise ratio (or equivalently, under Assumption 1).
dt
t dn
A dt
t t dr
A
t n t n t m k t v
Q
c c
d d f
) (
2
1 )] ( sin[ ) (
2
1
) ( where ), ( ) ( ) (
π
ψ
π
= = + ≈
) ( ) ( 8. : 6.2 Table f fG j t g
dt
d
π 2 ↔
) (t n
Q
c
A
f j
f H
π
π
2
2
) ( =
) (t n
d
80
© PoNing Chen@cm.nctu Chapter 2159
) ( ) (
) (
) ( ) ( ) ( ) (
2
2
f S
A
f
f S
A
f j
A
jf
f S f H f H f S
Q Q Q d
N
c
N
c c
N N
=
−
= − = ⇒
¦
¹
¦
´
¦
≤
= ⇒
otherwise , 0
2
  ,
) (
0
2
2
T
c
N
B
f N
A
f
f S
d
Bandwidth W < B
T
/2
that is just enough
to pass m(t).
) ( ) ( ) ( t n t m k t v
o f o
+ ≈
¦
¹
¦
´
¦
≤
= ⇒
otherwise , 0
  ,
) (
0
2
2
W f N
A
f
f S
c
N
o
¦
¹
¦
´
¦
<
+ + −
=
otherwise , 0
2 /   for
), ( ) (
) (
T
c N c N
N
B f
f f S f f S
f S
Q
© PoNing Chen@cm.nctu Chapter 2160
2
3
0
2
2
0
2
3
2
)] ( [
c
W
W
c
o
A
W N
df f
A
N
t n E = = ⇒
∫
−
Observation from the above formula:
. the named is This power. noise Decreasing
2 / power carrier increasing system, FM an In
2
effect quieting noise ≡
c
A
81
© PoNing Chen@cm.nctu Chapter 2161
). ( provided ,
2
3
3
2
)] ( [
3
0
2 2
2
3
0
2 2
FM ,
t r A
W N
P k A
A
W N
t m E k
SNR
c
f c
c
f
O
>> = = ⇒
), ( ) ( ) ( As t n t m k t v
o f o
+ ≈
. ) ( 2 ) ( where )], ( 2 cos[ ) (
0
∫
= + =
t
f c c
d m k t t t f A t s τ τ π φ φ π
We next turn to SNR
C,FM
.
. 2 / is ) ( signal modulated the in power average
2
c
A t s ⇒
.
2
is bandwidth message the in power noise Average
0
0
WN df
N
W
W
=
∫
−
.
2
2 /
0
2
0
2
FM ,
W N
A
W N
A
SNR
c c
C
= = ⇒
) 1 ( 2
1
1 2
1
1 2 . 2
. Hence,
. ratio Deviation 1.
. increasing increasing , fixed For :
Carson ,
2
FM ,
FM ,
2 / 1
FM ,
FM ,
+ =

.

\

+ =

.

\

+ ∆ =
∝
∝
∆
=
⇔
D W
D
DW
D
f B
D
SNR
SNR
W
P k
W
f
D
SNR
SNR
B W
T
C
O
f
C
O
T
Remarks
© PoNing Chen@cm.nctu Chapter 2162
.
3
2
2
3
2
2
0
2
3
0
2 2
FM ,
FM ,
W
P k
WN
A
W N
P k A
SNR
SNR
f
c
f c
C
O
= = ⇒
 ) (  max t m k f
f
= ∆
82
© PoNing Chen@cm.nctu Chapter 2163
2.13 Summary
Specifically,
for high carriertonoise ratio ρ (equivalently to the
assumption made in Assumption 1), an increase in
transmission bandwidth B
T
provides a corresponding
quadratic increase in figure of merit of a FM system.
So, there is a tradeoff between B
T
and figure of merit.
Notably, figure of merit for an AM system is
nothing to do with B
T
.
© PoNing Chen@cm.nctu Chapter 2164
Example 2.5 SingleTone Modulation
m(t) = A
m
cos(2πf
m
t)
Then we can represent the figure of merit in terms of
modulation index (or deviation ratio) β as (cf. slide 263):
In order to make the figure of metric for an FM system to
be superior to that for an AM system with 100%
modulation, it requires:
.
2
3
2
3
) 2 / ( 3 3
2
2
2
2
2 2
2
2
FM ,
FM ,
β =
∆
= = = ⇒
W
f
W
A k
W
P k
SNR
SNR
m f f
C
O
471 . 0
3
2
3
1
2
3
2
= > ⇒ ≥ β β
83
© PoNing Chen@cm.nctu Chapter 2165
2.13 Capture Effect
Recall that in Assumption 1, we assume A
c
>> r(t).
This somehow hints that the noise suppression of an FM
modulation works well when the noise (or other unwanted
modulated signal that cannot be filtered out by the bandpass or
lowpass filters) is weaker than the desired FM signal.
What if the unwanted FM signal is stronger than the desired FM
signal.
The FM receiver will capture the unwanted FM signal!
What if the unwanted FM signal has nearly equal strength as the
desired FM signal.
The FM receiver will fluctuates back and forth between them!
© PoNing Chen@cm.nctu Chapter 2166
2.13 FM Threshold Effect
Recall that in Assumption 1, we assume A
c
>> r(t)
(equivalently, a high carriertonoise ratio) to simplify θ(t)
so that the next formula holds.
However, a further decrease of carriertonoise ratio will
break the FM receiver (from a clicking sound down to a
crackling sound).
As the same as the AM modulation, this is also named the
threshold effect.
.
2
3
3
0
2 2
FM ,
W N
P k A
SNR
f c
O
=
84
© PoNing Chen@cm.nctu Chapter 2167
2.13 FM Threshold Effect
Consider a simplified case with m(t) = 0 (no message
signal).
) ( )] ( cos[ ) ( 2
) ( ' ) ( )] ( sin[ ) ( ' )] ( cos[ ) ( ' ) (
)] ( cos[ ) (
)] ( sin[ ) (
tan
) ( ' ) ( 2
2 2
2
1
t r t t r A A
t t r t t r A t t t r A
dt
t t r A
t t r
d
t t v
c c
c c
c
+ +
+ +
=


.

\

)
`
¹
¹
´
¦
+
= =
−
ψ
ψ ψ ψ ψ
ψ
ψ
θ π
To facilitate the understanding of “clicking” sound effect,
we let r(t) = λ A
c
, a constant ratio of A
c
.
© PoNing Chen@cm.nctu Chapter 2168
2.13 FM Threshold Effect
2
)] ( cos[ 2 1
)] ( cos[
) ( ' ) ( ' ) ( 2
λ ψ λ
λ ψ
λψ θ π
+ +
+
= = ⇒
t
t
t t t v
) ( '
1 )] ( cos[ 2 1
)] ( cos[
) ( ' ) ( ' ) ( 2
2
t
t
t
t t t v ψ
λ
λ
λ ψ λ
λ ψ
λψ θ π
−
≈
+ +
+
= = ⇒
1 but 1 and , ) ( say, time, at the Then ≈ > ≈ λ λ π ψ t
Thus a sign change in ψ’(t) will cause a spike!
Notably, when λ = 0 (no noise), the output equals m(t)
= 0 as desired.
85
© PoNing Chen@cm.nctu Chapter 2169
2.13 FM Threshold Effect
2
)] sin( cos[ 2 1
)] sin( cos[
) cos( ) ( ' ) ( 2 ) sin( ) (
λ π λ
λ π
λπ θ π π ψ
+ +
+
= = ⇒ =
t
t
t t t v t t
05 . 1 = λ
05 . 0 = λ
5 = λ
© PoNing Chen@cm.nctu Chapter 2170
2.13 Numerical Experiment
ratio. noise carrier to the is
2 /
and
2
,
2
3
and
2
,
4
3
) ( provided ,
2
3
0
2
3
2
3
0
2 2
3
0
2 2
FM ,
N B
A B
f
W
B
A k f
A
P
W N
f A
t r A
W N
P k A
SNR
T
c T T
m f
m c
c
f c
O
= = ∆

.

\

=
= ∆ =
∆
=
>> =
ρ ρ
Take B
T
/(2W) = 5.
The average output signal power is calculated in the
absence of noise.
The average output noise power is calculated when
there is no signal present.
86
© PoNing Chen@cm.nctu Chapter 2171
2.13 Numerical of the
formula in slide 2156
Curve I: The average noise
power is calculated assuming an
unmodulated carrier.
Curve II: The average noise
power is calculated assuming a
sinusoidally modulated carrier.
As text mentioned, for ρ <
11 dB, the output signalto
noise ratio deviates
appreciably from the linear
curve.
A true experiment found that
occasional clicks are heard at
ρ around 13 dB, only
slightly larger than what
theory indicates.
The deviation of
curve II from
curve I shows that
the average noise
power is indeed
dependent on the
modulating signal.
© PoNing Chen@cm.nctu Chapter 2172
2.13 How to avoid “clicking” sound?
For given modulation index (or deviation ratio) β and message
signal bandwidth W,
1. Determine B
T
by Carson’s rule or Universal curve.
2. For a specified noise level N
0
, select A
c
to satisfy:
. 20
2
y, equvalentl or dB 13
2
log 10
0
2
0
2
10
≥ ≥


.

\

N B
A
N B
A
T
c
T
c
87
© PoNing Chen@cm.nctu Chapter 2173
2.13 Threshold Reduction
After our learning
that FM
modulation has
threshold effect,
the next question
is naturally on
“how to reduce
the threshold?”
© PoNing Chen@cm.nctu Chapter 2174
2.13 Threshold Reduction
Threshold reduction in FM receivers may be
achieved by
1. negative feedback (commonly referred to as an
FMFB demodulator), or
2. phaselocked loop demodulator.
Why PLL can reduce threshold effect is not covered in
this course. Notably, the PLL system analysis in
Section 2.14 assumes noisefree transmission.
88
) (t v ) (t x
. ) ( 2 ) ( where )], ( 2 cos[ ) (
0
∫
= + =
t
f c c
d m k t t t f A t s τ τ π φ φ π
. ) ( 2 ) ( where )], ( 2 cos[ 2 ) (
0
∫
= + =
t
f vco vco vco vco
d m k t t t f t s τ τ πα φ φ π
)] ( ) 1 ( ) ( 2 cos[
)] ( 2 cos[ )] ( 2 cos[ 2 ) ( ) (
Bandpass
t t f f A
t t f t t f A t s t s
vco c c
vco vco c c vco
φ α π
φ π φ π
− + − →
+ + =
). ( at centered , ) 1 ( as wide as
passband smaller a have ly conceptual can filter bandpass the Thus,
vco c T
f f B − −α
© PoNing Chen@cm.nctu Chapter 2175
. ) 1 ( deviation frequency new The
original new
f f ∆ − = ∆ α
noise. e input whit the as level noise
same the with white treated be can output Mixer at the noise The
)] ( 2 cos[ ) ( 2 ) ( ) (
⇒
+ = t t f t w t s t w
vco vco vco
φ π
© PoNing Chen@cm.nctu Chapter 2176
)
`
¹
¹
´
¦
− − +
− −
+ − =
− = = =
+ →
+ + − + =
−
)] ( ) 1 ( ) ( cos[ ) (
)] ( ) 1 ( ) ( sin[ ) (
tan ) ( ) 1 ( ) ( and
, ) 1 ( )] ( [ )] ( [ )] ( [ where
)], ( 2 cos[
)] ( 2 cos[ ) ( )] ( ) 1 ( 2 cos[ ) (
1
0
2 2 2
'
Limiter
'
t t t r A
t t t r
t t
N B t n E t n E t n E
t t f
t t f t r t t f A t x
c
T Q I
c
c c c
φ α ψ
φ α ψ
φ α θ
α
θ π
ψ π φ α π
y. probabilit higer with holds ) ( ) ( ) ( of condition the
same, the remains and smaller, is )] ( [ )] ( [ Since
2 2
2 2
t n t n t r A
A t n E t n E
Q I c
c Q I
+ = >>
=
Experiments show that an FMFB receiver is capable of realizing a
threshold extension on the order of 5~7 dB.
89
© PoNing Chen@cm.nctu Chapter 2177
2.13 Threshold Reduction of an FMFB
receiver
To sum up:
An FMFB demodulator is essentially a tracking
filter that can track only the slowly varying
frequency of a wideband FM signal (i.e., the
message signal part), and consequently it responds
only to a narrowband of noise centered about the
instantaneous carrier frequency.
© PoNing Chen@cm.nctu Chapter 2178
2.13 PreEmphasis and Deemphasis in FM
Recall that the noise PSD at the output shapes like a bowel.
So if we can place more signal power at low frequency, a
better noise performance should result.
90
© PoNing Chen@cm.nctu Chapter 2179
2.13 PreEmphasis and Deemphasis in FM
In order to produce an undistorted version of the original
message at the receiver output, we must have:
This relation guarantees the intactness of the message power.
Next, we need to find H
de
(f) such that the noise power is
optimally suppressed.
. for 1 ) ( ) ( W f W f H f H
de pe
≤ ≤ − =
© PoNing Chen@cm.nctu Chapter 2180
2.13 PreEmphasis and Deemphasis in FM
Under the assumption of high carriertonoise ratio, the
noise PSD at the deemphasis filter output is given by:
¦
¹
¦
´
¦
≤
=
otherwise , 0
  ,  ) ( 
) (  ) ( 
2
2
2
0
2
W f f H
A
f N
f S f H
de
c
N de
o
∫
−
= ⇒
W
W
de
c
df f H
A
f N
2
2
2
0
 ) (  power noise Average
91
© PoNing Chen@cm.nctu Chapter 2181
2.13 PreEmphasis and Deemphasis in FM
Since the message power remains the same, the
improvement factor of the output signaltonoise ratio
before and after pre/deemphasis is:
∫ ∫
∫
∫
∫
− −
−
−
−
= = =
W
W
de
W
W
de
W
W
W
W
de
c
W
W
c
df f H f
W
df f H f
df f
df f H
A
f N
df
A
f N
I
2 2
3
2 2
2
2
2
2
0
2
2
0
 ) (  3
2
 ) ( 
 ) ( 
© PoNing Chen@cm.nctu Chapter 2182
Example 2.6
Preemphasis filter Deemphasis filter
). 2 /( 1 where ,
) / ( 1
1
) ( and ) / ( 1 ) (
0
0
0
Cr f
f f j
f H f f j f H
de pe
π =
+
= + =
.   for 1 2 and Assume W f fCr r R ≤ << << π
92
© PoNing Chen@cm.nctu Chapter 2183
Example 2.6
)] / ( tan ) / [( 3
) / (
) / ( 1
3
2
 ) (  3
2
0
1
0
3
0
2
0
2
3
2 2
3
f W f W
f W
df
f f
f
W
df f H f
W
I
W
W
W
W
de
−
−
−
−
=
+
= = ⇒
∫
∫
With f
0
= 21 KHz and W= 15 KHz, we obtain I = 22 = 13 dB.
A significant noise performance improvement is therefore obtained.
© PoNing Chen@cm.nctu Chapter 2184
2.13 PreEmphasis and Deemphasis in FM
Final remarks:
The previous example uses a linear preemphasis
and deemphasis filter to improve the noise.
A nonlinear preemphasis and deemphasis filters
have been applied successfully to application like
tape recording. These techniques, known as Dolby
A, DolbyB, and DBX systems, use a combination
of filtering and dynamic range compression to
reduce the effects of noise.
93
© PoNing Chen@cm.nctu Chapter 2185
2.14 Computer Experiments: PhaseLocked
Loop
Phaselocked loop
. ) ( 2 ) ( where )], ( 2 sin[ ) (
0
1 1
∫
= + =
t
f c c
d m k t t t f A t s τ τ π φ φ π
. ) ( 2 ) ( where )], ( 2 cos[ ) (
0
2 2
∫
= + =
t
v c v
d v k t t t f A t r τ τ π φ φ π
m
k
×
© PoNing Chen@cm.nctu Chapter 2186
( )
)], ( sin[
2
)] ( ) ( sin[ )] ( ) ( 4 sin[
2
)] ( 2 cos[ )] ( 2 sin[
) ( ) ( ) (
Low Pass
2 1 2 1
2 1
t
A A k
t t t t t f
A A k
t t f A t t f A k
t r t s k t e
e
v c m
c
v c m
c v c c m
m
φ
φ φ φ φ π
φ π φ π
→
− + + + =
+ ⋅ + =
=
m
k
×
Loop filter = low pass filter + filter h(τ).
). ( ) ( ) ( where
2 1
t t t
e
φ φ φ − =
. ) ( )} ( { ) ( Also,
LowPass
∫
∞
∞ −
− = τ τ τ d t h e t v
94
© PoNing Chen@cm.nctu Chapter 2187
∫
− = − =
t
v e
d v k t t t t
0
1 2 1
) ( 2 ) ( ) ( ) ( ) ( τ τ π φ φ φ φ
∫
∫
∞
∞ −
∞
∞ −
− − =
− − =
− =
⇒
τ τ τ φ π
φ
τ τ τ π
φ
π
φ φ
d t h k
dt
t d
d t h e k
dt
t d
t v k
dt
t d
dt
t d
e
v
v
e
) ( )] ( sin[ 2
) (
) ( )} ( { 2
) (
) ( 2
) ( ) (
0
1
LowPass
1
1
. 2 / where
0 v c v m
A A k k k =
( )
( )du u h u k t
du ds s u h s k du
du
u d
du
du
u d
t
t
e
t
e
t
t
e
e
∫
∫ ∫ ∫
∫
− =
− − =
=
⇒
∞
∞ −
0
0 1
0
0
0
1
0
) ( * )] ( sin[ 2 ) (
) ( )] ( sin[ 2
) (
) (
) (
φ π φ
φ π
φ
φ
φ
© PoNing Chen@cm.nctu Chapter 2188
The previous formula suggests an equivalent analytical model for PLL.
When φ
e
(t) = 0, the system is said to be in phaselock.
In this case, φ
1
(t) = φ
2
(t) or equivalently, k
v
v(t) = k
f
m(t).
When φ
e
(t) is small (< 0.5 radians), the system is said to be in near
phaselocked.
In this case, we can approximate sin[φ
e
(t)] by φ
e
(t); hence, a linear
approximate model is resulted.
∫
95
© PoNing Chen@cm.nctu Chapter 2189
Linearization approximation model for PLL.
∫
We can transform the above timedomain system to its
equivalent frequency domain to facilitate its analysis.
© PoNing Chen@cm.nctu Chapter 2190
Linearization approximation model for PLL.
) ( ) ( ) ( 2 1
1
) ( ) ( ) ( 2 ) (
) (
) ( )] ( ) ( [
) (
) (
) (
0 0
0
2 2 1 1
f H k jf
jf
f G f H k
f G f H f k f
f
f f f
f
f
f
e e
e
e e
+
=
+
=
Φ + Φ
Φ
=
Φ + Φ − Φ
Φ
=
Φ
Φ
π
π
) (
2
f Φ
) (
1
f Φ
) ( f H
) ( f G
f j
f G
π 2
1
) ( =
) ( f
e
Φ
96
© PoNing Chen@cm.nctu Chapter 2191
2.14 Experiment 0: FirstOrder PLL
. 1 ) ( = f H
) / ( 1
) / (
) (
) (
0
0
1
k f j
k f j
f
f
e
+
=
Φ
Φ
A parameter k
0
controls both the loop gain and bandwidth of the
filter. In other words, there is no why of adjusting loop gain
without changing the filter bandwidth.
© PoNing Chen@cm.nctu Chapter 2192
2.14 Experiment 1: SecondOrder
Acquisition PLL
model. linear PLL using and ) /( 1 ) ( jf a f H + =
2
2
0 0
2
2
0 0 1
) / ( ) / ( 2 1
) / (
) ( ) (
) (
)) /( 1 ( ) ( ) (
) (
n n
n
e
f jf f jf
f jf
a k jf k jf
jf
jf a k jf
jf
f H k jf
jf
f
f
+ +
=
+ +
=
+ +
=
+
=
Φ
Φ
ζ
. ) 4 /( factor damping and frequency natural where
0 0
a k ak f
n
= = ζ
97
© PoNing Chen@cm.nctu Chapter 2193
2.14 Experiment 1: SecondOrder
Acquisition PLL
Herz. ) 2 /( 1 Take π =
n
f
Fast response but require
longer time to stablelize
Slow response but quick
stablelizatoin.
© PoNing Chen@cm.nctu Chapter 2194
2.14 Experiment 2: PhasePlain Portrait
model. PLL nonlinear using and ) /( 1 ) ( jf a f H + =
¦
¦
¹
¦
¦
´
¦
=
=
damping) (critical
2
1
Herz
2 2
50
Take
ζ
π
n
f
Hz.
2 2
50
where ), cos(2 ) ( Let
π π
π = =
m m m
f t f A t m
∫
∫
∞
∞ −
∞
∞ −
− − =
− − =
τ τ τ φ π
τ τ τ φ
φ
π
φ
π
d t h t f
k
A k
d t h
dt
t d
k dt
t d
k
e m
m f
e
e
) ( )] ( sin[ ) 2 cos(
) ( )] ( sin[
) (
2
1 ) (
2
1
0
1
0 0
98
PhasePlane Portrait
© PoNing Chen@cm.nctu Chapter 2195
) (t
e
φ
) (
2
1
0
t
dt
d
k
e
φ
π
Initial value
Remarks on the previous figure
Each curve corresponds to different initial frequency error
(i.e., boundary condition for the differential equation).
The phaseplane portrait is periodic in φ
e
(t) with period 2π,
but the phaseplane portrait is not periodic in dφ
e
(t)/dt.
There exists an initial frequency error (such as 2 in the
previous figure) at which a saddle line appears (i.e., the
solid line in the previous figure).
In certain cases, the PLL will ultimately reach an
equilibrium (stable) points at (0,0) or (2π,0).
A slightest perturbation to the “saddle line” causes it to shift
to the equilibrium points.
© PoNing Chen@cm.nctu Chapter 2196
99
© PoNing Chen@cm.nctu Chapter 2197
2.15 Summary and Discussion
Four types of AM modulations are introduced
(expensive) DSBC transmitter + (inexpensive) envelope
detector, which is good for applications like radio
broadcasting.
(less expensive) DSBSC transmitter + (more complex)
coherent detector, which is good for applications like limited
transmitterpower pointtopoint communication.
(less bandwidth) VSB transmitter + coherent detector, which
is good for applications like television signals and high speed
data.
(minimum transmission power/bandwidth) SSB transmitter +
coherent detector, which is perhaps only good for
applications whose message signals have an energy gap on
zero frequency.
© PoNing Chen@cm.nctu Chapter 2198
2.15 Summary and Discussion
FM modulation, a representative of Angle
Modulation
A nonlinear modulation process
Carson’s rule and universal curve on transmission
bandwidth
100
© PoNing Chen@cm.nctu Chapter 2199
2.15 Summary and
Discussion
Noise performance
I. Full AM (DSBC) with
100 percent modulation
II. Coherent DSBSC &
SSB
III. FM with β = 2 and 13
dB pre/deemphasis
improvement
IV. FM with β = 5 and 13
dB pre/deemphasis
improvement
Threshold effect
© PoNing Chen@cm.nctu Chapter 2200
2.15 Summary and Discussion
Normalized transmission bandwidth in the previous figure
16 8 1 2 B
n
FM(β=5) FM(β=2) SSB AM, DSBSC
f
B
f W
f B
W
B
B
T T T
n
∆
=
∆
∆
= = β
/
/
urve UniversalC ,
4 2× = 2 . 3 5× =
(Refer to slide 85.)
101
© PoNing Chen@cm.nctu Chapter 2201
2.15 Summary and Discussion
Observations from the figure
1. SSB modulation is optimum in noise performance
and bandwidth conservation in AM family
2. FM modulation improves the noise performance of
AM family at the expense of an excessive
transmission bandwidth.
3. Curves III an IV indicates FM modulation offers
the tradeoff between transmission bandwidth and
noise performance.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.