Sirindhorn International Institute of Technology

Thammasat University
School of Information, Computer and Communication Technology

ECS 332: Principles of Communications

(Fourier Transform and Communication Systems)
Prapun Suksompong, Ph.D.
prapun@siit.tu.ac.th
August 10, 2012

Communication systems are usually viewed and analyzed in frequency domain. This
note reviews some basic properties of Fourier transform and introduce basic communication
systems.

Contents
1 Introduction to communication systems 3

2 Frequency-Domain Analysis 5
2.1 Math background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Continuous-Time Fourier Transform . . . . . . . . . . . . . . . . . . . . . . 6

3 Modulation and Frequency Shifting 19

4 Amplitude modulation: DSB-SC 23

4.1 Double-sideband suppressed carrier (DSB-SC) modulation . . . . . . . . . 23
4.2 Fourier Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.3 Fourier series expansion for real valued function . . . . . . . . . . . . . . . 27
4.4 Producing the modulated signal . . . . . . . . . . . . . . . . . . . . . . . . 29

5 Quadrature Amplitude Modulation (QAM) 33

1
6 Amplitude modulation: AM 37

7 Angle Modulation: FM and PM 43

7.1 Instantaneous Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

A Trig Identities 47

2
 Sirindhorn International Institute of Technology
Thammasat University
School of Information, Computer and Communication Technology

ECS332 2012/1 Part I Dr.Prapun

1 Introduction to communication systems
1.1. Shannon’s insight [8]:
The fundamental problem of communication is that of reproducing
at one point either exactly or approximately a message selected at
another point.
Definition 1.2. Figure 1 [8] shows a commonly used model for a (single-
link or point-to-point) communication system. All information transmission
systems involve three major subsystems–a transmitter, the channel, and a
receiver.
(a) Information1 source: produce a message
• Messages may be categorized as analog (continuous) or digital
(discrete).
(b) Transmitter: operate on the message to create a signal which can be
sent through a channel
(c) Channel: the medium over which the signal, carrying the information
that composes the message, is sent
• All channels have one thing in common: the signal undergoes
degradation from transmitter to receiver.
1
The concept of information is central to communication. But information is a loaded word, implying
semantic and philosophical notions that defy precise definition. We avoid these difficulties by dealing
instead with the message, defined as the physical manifestation of information as produced by the source.
[3, p 2]

3
 ◦ Although this degradation may occur at any point of the com-
munication system block diagram, it is customarily associated
with the channel alone.
◦ This degradation often results from noise 2 and other unde-
sired signals or interference 3 but also may include other dis-
tortion 4 effects as well, such as fading signal levels, multiple
Basic elements of communication
transmission paths, and filtering.
(d) Receiver: transform the signal back into the message intended for
 Information source: produce a message
delivery
 Transmitter: operate on the message to create a signal
(e) Destination: a person or a machine, for whom or which the message
which can be sent through a channel
is intended

Transmitted Received
Message Message
Signal Signal
Information Channel
Transmitter Receiver Destination
Source

Noise Source

29
Figure 1: Schematic diagram of a general communication system

2
Noise refers to random and unpredictable electrical signals produced by natural processes both internal
and external to the system. [3, p 4]
3
Interference is contamination by extraneous signals from human sourcesother transmitters, power lines
and machinery, switching circuits, and so on. Interference occurs most often in radio systems whose
receiving antennas usually intercept several signals at the same time. [3, p 4]
4
Distortion is waveform perturbation caused by imperfect response of the system to the desired signal
itself. Unlike noise and interference, distortion disappears when the signal is turned off. If the channel has
a linear but distorting response, then distortion may be corrected, or at least reduced, with the help of
special filters called equalizers. [3, p 4]

4
2 Frequency-Domain Analysis
Electrical engineers live in the two worlds, so to speak, of time and frequency.
Frequency-domain analysis is an extremely valuable tool to the communi-
cations engineer, more so perhaps than to other systems analysts. Since the
communications engineer is concerned primarily with signal bandwidths and
signal locations in the frequency domain, rather than with transient analysis,
the essentially steady-state approach of the (complex exponential) Fourier
series and transforms is used rather than the Laplace transform.

2.1 Math background

2.1. Euler’s formula: ejx = cos x + j sin x.

1 jA
cos (A) = Re ejA = e + e−jA


2
1 jA
e − e−jA .
 jA

sin (A) = Im e =
2j
2.2. We can use cos x = 12 ejx + e−jx and sin x = 2j1 ejx − e−jx to derive

many trigonometric identities.

Example 2.3. cos2 (x) = 21 (cos(2x) + 1)

2.4. Similar technique gives

5
(a) cos(−x) = cos(x),
(b) cos x − π2 = sin(x),

(c) sin(x) cos(x) = 12 sin(2x), and

(d) the product-to-sum formula
1
cos(x) cos(y) = (cos(x + y) + cos(x − y)) . (1)
2

2.2 Continuous-Time Fourier Transform

Definition 2.5. The (direct) Fourier transform of a signal g(t) is defined
by
Z+∞
G(f ) = g(t)e−j2πf t dt (2)
−∞

This provides the frequency-domain description of g(t). Conversion back to

the time domain is achieved via the inverse (Fourier) transform:
Z∞
g (t) = G (f ) ej2πf t df (3)
−∞

• We may combine (2) and (3) into one compact formula:

Z∞ Z∞
F
−−
G (f ) ej2πf t df = g (t) )−*
− G (f ) =
−1
g (t) e−j2πf t dt. (4)
F
−∞ −∞

• We may simply write G = F {g} and g = F −1 {G}.

R R
• Note that G(0) = g(t)dt and g(0) = G(f )df .
2.6. In some references5 , the (direct) Fourier transform of a signal g(t) is
defined by Z +∞
G2 (ω) = g(t)e−jωt dt (5)
−∞
5
MATLAB uses this definition.

6
In which case, we have
Z∞ Z∞
1 F
−−
G2 (ω) ejωt dω = g (t) ) −*
− G2 (ω) = g (t) e−jωt dt (6)
2π F −1

−∞ −∞
• In MATLAB, these calculations are carried out via the commands fourier
and ifourier.
1
R R
• Note that Ĝ(0) = g(t)dt and g(0) = 2π G(ω)dω.
• The relationship between G(f ) in (2) and G2 (ω) in (5) is given by
G(f ) = G2 (ω)|ω=2πf (7)
G2 (ω) = G(f )|f = ω (8)
2π

2.7. Q: The relationship between G(f ) in (2) and G2 (ω) in (5) is given by
(7) and (8) which do not involve a factor of 2π in the front. Why then does
1
the factor of 2π shows up in (6)?

Example 2.8. Rectangular and Sinc:

F sin(2πf a) 2 sin (aω)
−−
1 [|t| ≤ a] ) −*
− = = 2a sinc (aω) (9)
F −1 πf ω

7
 ⎛T ⎞
sinc ( 2π f 0t ) 1
1 ⎡ ω ≤ 2π f 0 ⎤⎦ T0sinc ⎜ 0 ω ⎟
2 f0 ⎣ ⎝2 ⎠
1 T0
1
2 f0 2π
1 T0
2 f0 ω ω
1
2π f 0
F
ZZZX
t YZZZ
F −1 t F
ZZZX
T0 T0 YZZZ
F −1 T0sinc (π T0 f )
1 −
1 2 2
2 f0 T0
f0 T0
f 1
T0
f0 f

Figure 2: Fourier transform of sinc and rectangular functions

• By setting a = T0 /2, we have

 
T0 F
1 |t| ≤ −−
)−*
− T0 sinc(πT0 f ). (10)
2 F −1

• In [4, p 78], the function 1 [|t| ≤ 0.5] is defined as the unit gate function
rect (x).

Definition 2.9. The function sinc(x) ≡ (sin x)/x is plotted in Figure 3.

1
1 𝑥

sinc(𝑥)

- 0  2 x

1
−
𝑥

Figure 3: Sinc function

• This function plays an important role in signal processing. It is also

known as the filtering or interpolating function.

8
 • Using L’Hôpital’s rule, we find lim sinc(x) = 1.
x→0

• sinc(x) is the product of an oscillating signal sin(x) (of period 2π) and
a monotonically decreasing function 1/x . Therefore, sinc(x) exhibits
sinusoidal oscillations of period 2π, with amplitude decreasing contin-
uously as 1/x.
• In MATLAB and in [10, eq. 2.64], sinc(x) is defined as (sin(πx))/πx. In
which case, it is an even damped oscillatory function with zero crossings
at integer values of its argument.

Definition 2.10. The (Dirac) delta function or (unit) impulse function

is denoted by δ(t). It is usually depicted as a vertical arrow at the origin.
Note that δ(t) is not a true function; it is undefined at t = 0. We define
δ(t) as a generalized function which satisfies the sampling property (or
sifting property) Z ∞
φ(t)δ(t)dt = φ(0) (11)
−∞
for any function φ(t) which is continuous at t = 0.
• In this way, the delta “function” has no mathematical or physical mean-
ing unless it appears under the operation of integration.
• Intuitively we may visualize δ(t) as an infinitely
 tall, infinitely narrow
1 ε
rectangular pulse of unit area: lim ε 1 |t| ≤ 2 .
ε→0

2.11. Properties of δ(t):

• δ(t) = 0 for t 6= 0.
δ(t − T ) = 0 for t 6= T .
R
• A δ(t)dt = 1A (0).
R
(a) δ(t)dt = 1.
R
(b) {0} δ(t)dt = 1.
Rx
(c) −∞ δ(t)dt = 1[0,∞) (x). Hence, we may think of δ(t) as the “deriva-
tive” of the unit step function U (t) = 1[0,∞) (x).

9
 R
• φ(t)δ(t − c)dt = φ(c) for φ continuous at T . In fact, for any ε > 0,
Z T +ε
φ(t)δ(t − c)dt = φ(c).
T −ε

• Convolution property:
Z ∞
(δ ∗ φ)(t) = (φ ∗ δ)(t) = φ(τ )δ(t − τ )dτ = φ(t) (12)
−∞

where we assume that φ is continuous at t.

1
• δ(at) = |a| δ(t). In particular,

1
δ(ω) = δ(f ) (13)
2π
and
1
δ(ω − ω0 ) = δ(2πf − 2πf0 ) = δ(f − f0 ), (14)
2π
where ω = 2πf and ω0 = 2πf0 .
F
−−
Example 2.12. δ(t) )−*
− 1.
−1 F

F
−−
Example 2.13. ej2πf0 t )−*
− δ (f − f0 ).
−1 F

F
−−
Example 2.14. ejω0 t )−*
− 2πδ (ω − ω0 ).
−1 F

10
 F
Example 2.15. cos(2πf0 t) −
)−
−− 1 (δ (f − f0 ) + δ (f + f0 )).
*
−1 2 F

2.16. Conjugate symmetry6 : If x(t) is real-valued, then X(−f ) =

(X(f ))∗

Observe that if we know X(f ) for all f positive, we also know X(f ) for
all f negative. Interpretation: Only half of the spectrum contains all of
the information. Positive-frequency part of the spectrum contains all the
necessary information. The negative-frequency half of the spectrum can be
determined by simply complex conjugating the positive-frequency half of
the spectrum.
2.17. Shifting properties
• Time-shift:
F
g (t − t1 ) −
)−
−− e−j2πf t1 G (f )
*
−1
F

◦ Note that |e−j2πf t1 | = 1. So, the spectrum of g (t − t1 ) looks exactly

the same as the spectrum of g(t) (unless you also look at their
phases).

• Frequency-shift (or modulation):

F
ej2πf1 t g (t) −
)−
−*
− G (f − f1 )
−1
F
6
Hermitian symmetry in [7, p 17 ].

11
 2.18. Let g(t), g1 (t), and g2 (t) denote signals with G(f ), G1 (f ), and G2 (f )
denoting their respective Fourier transforms.
(a) Superposition theorem (linearity):
F
a1 g1 (t) + a2 g2 (t) −
)−
−*
− a1 G1 (f ) + a2 G2 (f ).
−1
F

(b) Scale-change theorem (scaling property [4, p 88]):

 
F 1 f
g(at) −)−
−*
− G .
F −1 |a| a
• The function g(at) represents the function g(t) compressed in time
by a factor a (when |a| > l). Similarly, the function G(f /a) repre-
sents the function G(f ) expand ed in frequency by the same factor
a.
• The scaling property says that if we “squeeze” a function in t,
its Fourier transform “stretches out” in f . It is not possible to
arbitrarily concentrate both a function and its Fourier transform.
• Generally speaking, the more concentrated g(t) is, the more spread
out its Fourier transform G(f ) must be.
• This trade-off can be formalized in the form of an uncertainty prin-
ciple. See also 2.28 and 2.29.
• Intuitively, we understand that compression in time by a factor
a means that the signal is varying more rapidly by the same fac-
tor. To synthesize such a signal, the frequencies of its sinusoidal
components must be increased by the factor a, implying that its
frequency spectrum is expanded by the factor a. Similarly, a signal
expanded in time varies more slowly; hence, the frequencies of its
components are lowered, implying that its frequency spectrum is
compressed.

12
 (c) Duality theorem (Symmetry Property [4, p 86]):
F
G(t) −
)−
−*
− g(−f ).
−1 F

• In words, for any result or relationship between g(t) and G(f ),

there exists a dual result or relationship, obtained by interchanging
the roles of g(t) and G(f ) in the original result (along with some
minor modifications arising because of a sign change).
In particular, if the Fourier transform of g(t) is G(f ), then the
Fourier transform of G(f ) with f replaced by t is the original time-
domain signal with t replaced by −f .
• If we use the ω-definition (5), we get a similar relationship with an
extra factor of 2π:
F
−−
G2 (t) )−*
− 2πg(−ω).
−1 F

F
−−
Example 2.19. x(t) = cos(2πaf0 t) )−− 1 (δ(f − af0 ) + δ(f + af0 )) .
*
−1 2 F

Example 2.20. From Example 2.8, we know that

F
1 [|t| ≤ a] −
)−
−*
− 2a sinc (2πaf )
−1
(15)
F

By the duality theorem, we have

F
2a sinc(2πat) −
)−
−*
− 1[| − f | ≤ a],
−1
F

which is the same as

F 1
−−
sinc(2πf0 t) )−*
− 1[|f | ≤ f0 ]. (16)
F −1 2f0
Both transform pairs are illustrated in Figure 2.

13
Example 2.21. Let’s try to derive the time-shift property from the frequency-
shift property. We start with an arbitrary function g(t). Next we will define
another function x(t) by setting X(f ) to be g(f ). Note that f here is just
a dummy variable; we can also write X(t) = g(t). Applying the duality
F
theorem to the transform pair x(t) ) −−
−*
− X(f ), we get another transform
F −1
F
pair X(t) −
)−
−*
− x(−f ). The LHS is g(t); therefore, the RHS must be G(f ).
−1
F
This implies G(f ) = x(−f ). Next, recall the frequency-shift property:
F
−−
ej2πct x (t) )−*
− X (f − c) .
F−1

The duality theorem then gives

F
−−
X (t − c) )−− ej2πc−f x (−f ) .
*
−1
F

Replacing X(t) by g(t) and x(−f ) by G(f ), we finally get the time-shift
property.
Definition 2.22. The convolution of two signals, x1 (t) and x2 (t), is a new
function of time, x(t). We write

x = x1 ∗ x2 .

It is defined as the integral of the product of the two functions after one is
reversed and shifted:

x(t) = (x1 ∗ x2 )(t) (17)

Z +∞ Z +∞
= x1 (µ)x2 (t − µ)dµ = x1 (t − µ)x2 (µ)dµ. (18)
−∞ −∞

• Note that t is a parameter as far as the integration is concerned.

• The integrand is formed from x1 and x2 by three operations:
(a) time reversal to obtain x2 (−µ),
(b) time shifting to obtain x2 (−(µ − t)) = x2 (t − µ), and
(c) multiplication of x1 (µ) and x2 (t − µ) to form the integrand.
• In some references, (17) is expressed as x(t) = x1 (t) ∗ x2 (t).

14
Example 2.23. We can get a triangle from convolution of two rectangular
waves. In particular,

1[|t| ≤ a] ∗ 1[|t| ≤ a] = (2a − |t|) × 1[|t| ≤ 2a].

2.24. Convolution theorem:

(a) Convolution-in-time rule:
F
−−
x1 ∗ x2 )−*
− X 1 × X2 . (19)
F−1

(b) Convolution-in-frequency rule:

F
x1 × x2 −
)−
−*
− X 1 ∗ X2 .
−1
(20)
F

Example 2.25. We can use the convolution theorem to “prove” the frequency-
sift property in 2.17.

2.26. From the convolution theorem, we have

F
−−
• g2 )−*
−G∗G
F −1

• if g is band-limited to B, then g 2 is band-limited to 2B

15
 2.27. Parseval’s theorem (Rayleigh’s energy theorem, Plancherel for-
mula) for Fourier transform:
Z +∞ Z +∞
|g(t)|2 dt = |G(f )|2 df. (21)
−∞ −∞

The LHS of (21) is called the (total) energy of g(t). On the RHS, |G(f )|2
is called the energy spectral density of g(t). By integrating the energy
spectral density over all frequency, we obtain the signal ’s total energy. The
energy
R contained in the frequency band B can be found from the integral
2
B |G(f )| df .
More generally, Fourier transform preserves the inner product [2, Theo-
rem 2.12]:
Z ∞ Z ∞
∗
hg1 , g2 i = g1 (t)g2 (t)dt = G1 (f )G∗2 (f )df = hG1 , G2 i.
−∞ −∞

2.28. (Heisenberg) Uncertainty Principle [2, 9]: RSuppose g is a func-

tion which satisfies the normalizingRcondition kgk22 = |g(t)|2 dt = 1 which
automatically implies that kGk22 = |G(f )|2 df = 1. Then
Z  Z 
1
t2 |g(t)|2 dt f 2 |G(f )|2 df ≥ , (22)
16π 2
2
−Bt
and
p equality holds if and only if g(t) = Ae where B > 0 and |A|2 =
2B/π.
• In fact, we have
Z  Z 
1
t2 |g(t − t0 )|2 dt f 2 |G(f − f0 )|2 df ≥ ,
16π 2
for every t0 , f0 .
• The proof relies on Cauchy-Schwarz inequality.
R 2 2
• For any function h, define its dispersion ∆h as Rt |h(t)| dt . Then, we can
2
|h(t)| dt
apply (22) to the function g(t) = h(t)/khk2 and get
1
∆h × ∆H ≥ .
16π 2

16
2.29. A signal cannot be simultaneously time-limited and band-limited.
Proof. Suppose g(t) is simultaneously (1) time-limited to T0 and (2) band-
limited to B. Pick any positive number Ts and positive integer K such that
fs = T1s > 2B and K > TT0s . The sampled signal gTs (t) is given by

X K
X
gTs (t) = g[k]δ (t − kTs ) = g[k]δ (t − kTs )
k k=−K

where g[k] = g (kTs ). Now, because we sample the signal faster than the
Nyquist rate, we can reconstruct the signal g by producing gTs ∗ hr where
the LPF hr is given by

Hr (ω) = Ts 1[ω < 2πfc ]

1
with the restriction that B < fc < Ts − B. In frequency domain, we have
K
X
G(ω) = g[k]e−jkωTs Hr (ω).
k=−K

Consider ω inside the interval I = (2πB, 2πfc ). Then,

K K
ω<2πfc jωTs
ω>2πB X
−jkωTs z=e=
X
0 = G(ω) = Ts g (kTs ) e Ts g (kTs ) z −k
k=−K k=−K
(23)
−K
Because z 6= 0, we can divide (23) by z and then the last term becomes
a polynomial of the form

a2K z 2K + a2K−1 z 2K−1 + · · · + a1 z + a0 .

By fundamental theorem of algebra, this polynomial has only finitely many

roots– that is there are only finitely many values of z = ejωTs which satisfies
(23). Because there are uncountably many values of ω in the interval I and
hence uncountably many values of z = ejωTs which satisfy (23), we have a
contradiction.
2.30. The observation in 2.29 raises concerns about the signal and filter
models used in the study of communication systems. Since a signal cannot
be both bandlimited and timelimited, we should either abandon bandlimited

17
signals (and ideal filters) or else accept signal models that exist for all time.
On the one hand, we recognize that any real signal is timelimited, having
starting and ending times. On the other hand, the concepts of bandlimited
spectra and ideal filters are too useful and appealing to be dismissed entirely.
The resolution of our dilemma is really not so difficult, requiring but a
small compromise. Although a strictly timelimited signal is not strictly ban-
dlimited, its spectrum may be negligibly small above some upper frequency
limit B. Likewise, a strictly bandlimited signal may be negligibly small out-
side a certain time interval t1 ≤ t ≤ t2 . Therefore, we will often assume that
signals are essentially both bandlimited and timelimited for most practical
purposes.

18
 Sirindhorn International Institute of Technology
Thammasat University
School of Information, Computer and Communication Technology

ECS332 2012/1 Part II.1 Dr.Prapun

3 Modulation and Frequency Shifting
Definition 3.1. The term baseband is used to designate the band of fre-
quencies of the signal delivered by the source.
Example 3.2. In telephony, the baseband is the audio band (band of voice
signals) of 0 to 3.5 kHz.
Example 3.3. For digital data (sequence of two voltage levels representing
0 and 1) at a rate of R bits per second, the baseband is 0 to R Hz.
Definition 3.4. Modulation is a process that causes a shift in the range
of frequencies in a signal.
• The modulation process commonly translates an information-bearing
signal to a new spectral location depending upon the intended fre-
quency for transmission.
Definition 3.5. In baseband communication, baseband signals are trans-
mitted without modulation, that is, without any shift in the range of fre-
quencies of the signal.
3.6. Recall the frequency-shift property:
F
−−
ej2πfc t g (t) )−*
− G (f − fc ) .
F −1

This property states that multiplication of a signal by a factor ej2πfc t shifts

the spectrum of that signal by ∆f = fc .

19
 3.7. Frequency-shifting (frequency translation) in practice is achieved by
multiplying g(t) by a sinusoid:
F 1
g(t) cos(2πfc t) −
)−
−*
− (G(f − fc ) + G(f + fc )) .
−1F 2

−2

−4
0 5 10 15 20 25 30

−2

−4
0 5 10 15 20 25 30

−2

−4
0 5 10 15 20 25 30

Similarly,
F 1
−−* G(f − fc )ejφ + G(f + fc )e−jφ .


g(t) cos(2πfc t + φ) )−−
−1
F 2

20
Definition 3.8. cos(2πfc t + φ) is called the (sinusoidal) carrier signal and
fc is called the carrier frequency. In general, it can also has amplitude A
and hence the general expression of the carrier signal is A cos(2πfc t + φ).
3.9. Examples of situations where modulation (spectrum shifting) is useful:
(a) Channel passband matching: Recall that, for a linear, time-invariant
(LTI) system, the input-output relationship is given by
y(t) = h(t) ∗ x(t)
where x(t) is the input, y(t) is the output, and h(t) is the impulse
response of the system. In which case,
Y (f ) = H(f )X(f )
where H(f ) is called the transfer function or frequency response
of the system. |H(f )| and ∠H(f ) are called the amplitude response
and phase response, respectively. Their plots as functions of f show
at a glance how the system modifies the amplitudes and phases of
various sinusoidal inputs.

(b) Reasonable antenna size: For effective radiation of power over a

radio link, the antenna size must be on the order of the wavelength of
the signal to be radiated.
• Audio signal frequencies are so low (wavelengths are so large) that
impracticably large antennas will be required for radiation. Here,

21
 shifting the spectrum to a higher frequency (a smaller wavelength)
by modulation solves the problem.

(c) Frequency-Division Multiplexing (FDM) and Frequency-Division

Multiple Access (FDMA):
• If several signals, each occupying the same frequency band, are
transmitted simultaneously over the same transmission medium,
they will all interfere; it will be difficult to separate or retrieve
them at a receiver.
• For example, if all radio stations decide to broadcast audio signals
simultaneously, the receiver will not be able to separate them.
• One solution is to use modulation whereby each radio station is as-
signed a distinct carrier frequency. Each station transmits a modu-
lated signal, thus shifting the signal spectrum to its allocated band,
which is not occupied by any other station. A radio receiver can
pick up any station by tuning to the band of the desired station.
Definition 3.10. Communication that uses modulation to shift the fre-
quency spectrum of a signal is known as carrier communication. [4, p
151]
3.11. A sinusoidal carrier signal A cos(2πfc t+φ) has three basic parameters:
amplitude, frequency, and phase. Varying these parameters in proportion
to the baseband signal results in amplitude modulation (AM), frequency
modulation (FM), and phase modulation (PM), respectively. Collectively,
these techniques are called continuous-wave modulation in [10, p 111].
We will use m(t) to denote the baseband signal. We will assume that
m(t) is band-limited to B; that is, |M (f )| = 0 for |f | > B. Note that we
usually call it the message or the modulating signal.
Definition 3.12. The process of recovering the signal from the modulated
signal (retranslating the spectrum to its original position) is referred to as
demodulation, or detection.

22
4 Amplitude modulation: DSB-SC
Definition 4.1. Amplitude modulation is characterized by the fact that
the amplitude A of the carrier A cos(2πfc t + φ) is varied in proportion to
the baseband (message) signal m(t).
• Because the amplitude is time-varying, we may write the modulated
carrier as
A(t) cos(2πfc t + φ)

• Because the amplitude is linearly related to the message signal, this

technique is also called linear modulation.

4.1 Double-sideband suppressed carrier (DSB-SC) modulation

4.2. Basic idea:
 
 
√  √

 

LPF m (t) × 2 cos (2πfc t) × 2 cos (2πfc t) = m (t) . (24)

 

| {z } 
x(t)

√ √
x (t) = m (t) × 2 cos (2πfc t) = 2m (t) cos (2πfc t)
√ 1
 
X (f ) = 2 (M (f − fc ) + M (f + fc ))
2
1
= √ (M (f − fc ) + M (f + fc ))
2
Similarly,
√ √
v (t) = y (t) × 2 cos (2πfc t) = 2x (t) cos (2πfc t)
1
V (f ) = √ (X (f − fc ) + X (f + fc ))
2

23
 Alternatively, we can use the trig. identity from Example 2.3:
√ √ √ 
v (t) = 2x (t) cos (2πfc t) = 2 2m (t) cos (2πfc t) cos (2πfc t)
= 2m (t) cos2 (2πfc t) = m (t) (cos (2 (2πfc t)) + 1)
= m (t) + m (t) cos (2π (2fc ) t)

4.3. In the process of modulation, observe that we need fc > B in order to

avoid overlap of the spectra.
4.4. Observe that the modulated signal spectrum centered at fc , is com-
posed of two parts: a portion that lies above fc , known as the upper side-
band (USB), and a portion that lies below fc , known as the lower side-
band (LSB). Similarly, the spectrum centered at −fc has upper and lower
sidebands. Hence, this is a modulation scheme with double sidebands.
4.5. Observe that (24) requires that we can generate cos (ωc t) both at the
transmitter and receiver. This can be difficult in practice. Suppose that
the frequency at the receiver is off, say by ∆f , and that the phase is off
by θ. The effect of these frequency and phase offsets can be quantified by
calculating
n √ √ o
LPF m (t) 2 cos ωc t 2 cos ((ωc + ∆ω) t + θ) ,
which gives
m (t) cos ((∆ω) t + θ) .
Of course, we want ∆ω = 0 and θ = 0; that is the receiver must generate
a carrier in phase and frequency synchronism with the incoming carrier.
These demodulators are called synchronous or coherent (also homo-
dyne) demodulator [4, p 161].
4.6. Effect of time delay: Suppose the propagation time is τ , then we
have √
y (t) = x (t − τ ) = 2m (t − τ ) cos (2πfc (t − τ ))
√
= 2m (t − τ ) cos (2πfc t − 2πfc τ )
√
= 2m (t − τ ) cos (2πfc t − φτ ) .

24
Consequently,
√
v (t) = y (t) × 2 cos (2πfc t)
√ √
= 2m (t − τ ) cos (2πfc t − φτ ) × 2 cos (2πfc t)
= m (t − τ ) 2 cos (2πfc t − φτ ) cos (2πfc t) .
Applying the product-to-sum formula, we then have

v (t) = m (t − τ ) (cos (2π (2fc ) t − φτ ) + cos (φτ )) .

4.2 Fourier Series

Let the (real or complex) signal r (t) be a periodic signal with period T0 .
Suppose the following Dirichlet conditions are satisfied
RT0
(a) r (t) is absolutely integrable over its period; i.e., |r (t)|dt < ∞.
0

(b) The number of maxima and minima of r (t) in each period is finite.
(c) The number of discontinuities of r (t) in each period is finite.
Then r
(t) can be expanded in terms of the complex exponential signals
∞
ejnω0 t n=−∞ as
∞
X ∞
X
jnω0 t
ck ejkω0 t + c−k e−jkω0 t


r̃ (t) = cn e = c0 + (25)
n=−∞ k=1

where
2π
ω0 = 2πf0 = ,
T0
α+T0
Z
1
ck = r (t) e−jkω0 t dt, (26)
T0
α

25
for some arbitrary α. In which case,

r (t) , if r (t) is continuous at t
r̃ (t) = r(t+ )+r(t− )
2 , if r (t) is not continuous at t
We give some remarks here.
• The parameter α in the limits of the integration (26) is arbitrary. It
can be chosen to simplify
R computation of the integral. Some references
1 −jkω0 t
simply write ck = T0 r (t) e dt to emphasize that we only need
T0
to integrate over one period of the signal; the starting point is not
important.
• The coefficients ck = T10 r (t) e−jkω0 t dt are called the (k th ) Fourier
R
T0
(series) coefficients of (the signal) r (t). These are, in general, com-
plex numbers.
• c0 = T10 r (t) dt = average or DC value of r(t)
R
T0

• The quantity f0 = T10 is called the fundamental frequency of the

signal r (t). The nth multiple of the fundamental frequency (for positive
n’s) is called the nth harmonic.
• ck ejkω0 t + c−k e−jkω0 t = the k th harmonic component of r (t).
k = 1 ⇒ fundamental component of r (t).
4.7. Consider a restricted version rT0 (t) of r(t) where we only consider r(t)
F
−−
for one specific period. Suppose rT0 (t) ) −*
− RT0 (f ). Then,
F −1

1
ck = RT (kf0 ).
T0 0
So, the Fourier coefficients are simply scaled samples of the Fourier trans-
form.
∞
2
1
|ck |2
R P
4.8. Parseval’s Identity: Pr = T0 |r (t)| dt =
T0 k=−∞

26
4.3 Fourier series expansion for real valued function
4.9. Suppose r (t) in the previous section is real-valued; that is r∗ = r.
Then, we have c−k = c∗k and we provide here three alternative ways to
represent the Fourier series expansion:
∞
X ∞
X
jnω0 t
ck ejkω0 t + c−k e−jkω0 t


r̃ (t) = cn e = c0 + (27)
n=−∞ k=1
∞
X ∞
X
= c0 + (ak cos (kω0 t)) + (bk sin (kω0 t)) (28)
k=1 k=1
X∞
= c0 + 2 |ck | cos (kω0 t + ∠ck ) (29)
k=1

where the corresponding coefficients are obtained from

α+T0
Z
1 1
ck = r (t) e−jkω0 t dt = (ak − jbk ) (30)
T0 2
α
Z
2
ak = 2Re {ck } = r (t) cos (kω0 t) dt (31)
T0
T0
Z
2
bk = −2Im {ck } = r (t) sin (kω0 t) dt (32)
T0
T0
q
2 |ck | = a2k + b2k (33)
 
bk
∠ck = − arctan (34)
ak
a0
c0 = (35)
2
The Parseval’s identity can then be expressed as
Z ∞ ∞
1 2
X 2
X
Pr = |r (t)| dt = 2
|ck | = c0 + 2 |ck |2
T0
T0 k=−∞ k=1

27
 4.10. To go from (27) to (28) and (29), note that when we replace c−k by
c∗k , we have
ck ejkω0 t + c−k e−jkω0 t = ck ejkω0 t + c∗k e−jkω0 t

∗
= ck ejkω0 t + ck ejkω0 t
= 2 Re ck ejkω0 t .


• Expression (29) then follows directly from the phasor concept:

Re ck ejkω0 t = |ck | cos (kω0 t + ∠ck ) .


• To get (28), substitute ck by Re {ck } + j Im {ck }

and ejkω0 t by cos (kω0 t) + j sin (kω0 t).
Example 4.11. Train of impulses:
∞ ∞ ∞
(T0 )
X 1 X jkω0 t 1 2 X
δ (t) = δ (t − kT0 ) = e = + cos kω0 t (36)
T0 T0 T0
k=−∞ k=−∞ k=1

t
T0 2T0

Figure 4: Train of impulses

1
Example 4.12. Square pulse periodic signal:
 
1 2 −T 1 1T t 1
1 [cos ω0 t ≥ 0] = + cos ω0 t − cos 3ω0 t + cos 5ω0 t − cos 7ω0 t + . . .
0 0

2 π 3 -1
5 7
(37)
We note here that multiplication by this signal is a switching function.

−T0 T0 t

Figure 5: Square pulse periodic signal

28
Example 4.13. Bipolar square pulse periodic signal:
 
4 1 1 1
sgn(cos ω0 t) = cos ω0 t − cos 3ω0 t + cos 5ω0 t − cos 7ω0 t + . . .
π 3 5 7

−T0 T0 t

-1

Figure 6: Bipolar square pulse periodic signal

4.4 Producing the modulated signal

−T0 T0 t
To produce the modulated signal m(t) cos(2πfc t), we may use the following
methods which generate the modulated signal along with other signals which
can be eliminated by a bandpass filter restricting frequency contents to
around ωc .
4.14. Multiplier Modulators: Here modulation is achieved directly by
multiplying m(t) by cos(2πfc t) using an analog multiplier whose output is
proportional to the product of two input signals.
• Such a multiplier may be obtained from a variable-gain amplifier in
which the gain parameter (such as the the β of a transistor) is con-
trolled by one of the signals, say, m(t). When the signal cos(2πfc t) is
applied at the input of this amplifier, the output is then proportional
to m(t) cos(2πfc t).
• Another way to multiply two signals is through logarithmic amplifiers.
Here, the basic components are a logarithmic and an antilogarithmic
amplifier with outputs proportional to the log and antilog of their in-
puts, respectively. Using two logarithmic amplifiers, we generate and
add the logarithms of the two signals to be multiplied. The sum is then
applied to an antilogarithmic amplifier to obtain the desired product.
• Difficult to maintain linearity in this kind of amplifier.

29
 • Expensive.
4.15. Square Modulator: When it is easier to build a squarer than a
multiplier, use
(m (t) + c cos (ωc t))2 = m2 (t) + 2c m (t) cos (ωc t) + c2 cos2 (ωc t)
2 c2 c2
= m (t) + +2c m (t) cos (ωc t) + + cos (2ωc t) .
2 2

ωc


3
• Alternative, can use m(t) + c cos 2t .

4.16. Multiply m(t) by “any” periodic and even signal r(t) whose period
is Tc = 2π
ωc . Because r(t) is an even function, we know that
∞
X
r (t) = c0 + ak cos (kωc t).
k=1

Therefore,
∞
X
m(t)r (t) = c0 m(t) + ak m(t) cos (kωc t).
k=1

See also [4, p 157]. In general, for this scheme to work, we need
• a1 6= 0; that is Tc is the “least” period of r;
• ωc > 4πB; that is fc > 2B (to prevent overlapping).

30
 m (t )
× BPF m ( t ) cos (ωct )

r (t ) F {m × r}(ω )
M (ω ) 1 1
Aa Aa2
A c0 A 2 1
2

−2ωc −ωc ωc 2ωc

2π B
2π B
ωc − 2π B
BPF

Figure 7: Modulation of m(t) via even and periodic r(t)

Note that if r(t) is not even, then by (29), the outputted modulated
signal is of the form a1 m(t) cos(ωc t + φ1 ).
4.17. Switching modulator : Set r(t) to be the square pulse train given
by (37):

r (t) = 1 [cos ω0 t ≥ 0]
 
1 2 1 1 1
= + cos ω0 t − cos 3ω0 t + cos 5ω0 t − cos 7ω0 t + . . . .
2 π 3 5 7
Multiplying this r(t) to the signal m(t) is equivalent to switching m(t) on
and off periodically.
It is equivalent to periodically turning the switch on (letting m(t) pass
through) for half a period Tc = f1c .

31
186 AMPLITUDE MODULATIONS AND DEMODULATIONS

Figure 4.4
Switching m(t )

~
modulator for
DSB-SC.

0 J~
(a) I
w(l)

nnnnnnnnnnnnn
(b)

Figure 8: Switching
The square modulator
pulse train w(t) in Fig. 4.4bfor
is a DSB-SC [4, Figure
periodic signal 4.4]. series was found
whose Fourier
earlier in Example 2.8 [Eq. (2.86)] as
4.18. Switching Demodulator :
ll' (l) = ~2 + ~ (cos We t - ~ cos 3wct + ~ cos Seve! - ·· ·) (4 .S )
J[ 1 .) )

LPF{m(t) cos(ωc t) × 1[cos(ωc t) ≥ 0]} = m(t) (38)

T he signalm (t)lr (t) is given by
π
[4, p 162]. Note that this technique still requires the switching to be in sync
l +as-2 in
[ m (t) cos eve! - -m
with the incoming
m (t)w(t) cosine
= -m(t)
2 the basic 3IDSB-SC.
J[ )
l (1) cos Seve ! - · · · ] (4.6)
(!) cos 3wct + -::111

The signal m(l) ll '(t) co nsists not onl y of the component 111(1) bu t a lso of an infinite
number of modul ated signals with angul ar frequ encies r»c, 3wc, Seve, .. .. Therefore, the
spectrum of m (1)1-v(t) con sists of multiple copies of the message spectrum M (f), shifted to
0, ±fc, ±~fc , ±Sf~- , .. . (with decreas ing re lative weights), as show n in Fig. 4.4c.
For modulation , we are interested in extracting the modu lated co mponent m.(t) cos We i
on ly. To separate this component from the rest of the crowd, we pass the signal m(t)w(t) through
a bandpass filter of band width 28 H z (or 4Tr8 rad/s) , centered at the freque ncy ±fc · Pro vided the
carrier angular frequ ency.fc ::::: 28 (or We ::::: 4Tr8) , thi s will suppress all the spectral components
not centered at ±fc to yield the des ired modulated signal (2/ Tr)m (t) cos We t (Fig. 4.4d).
We now see the real payoff of this method. Multipl ication of a signal by a square pul se train
is in reality a sw itching operation. It involves switching the signal m(t) on and off periodically
and can be accomplished by simple switching elements controlled by w(t). Figure 4 .Sa shows
one such electronic sw itch, the diode bridge modulator, driven by a sinusoid A cos We t to
produce the switching acti on. Diodes D 1, D2 and D3, D4 are matched pairs. When the signal
cos Wet is of a polarity that wi ll make terminal c positive with respect to d, all the diodes

32
 Sirindhorn International Institute of Technology
Thammasat University
School of Information, Computer and Communication Technology

ECS332 2012/1 Part II.2 Dr.Prapun

5 Quadrature Amplitude Modulation (QAM)
Definition 5.1. One of the possible definition for the bandwidth (BW )
of a signal is the difference between the highest frequency and the lowest
frequency in the positive-f part of the signal spectrum.
Example 5.2.

5.3. Rough Approximation: If g1 (t) and g2 (t) have bandwidths B1 and B2

Hz, respectively, the bandwidth of g1 (t)g2 (t) is B1 + B2 Hz.
This result follows from the application of the width property7 of convo-
lution8 to the convolution-in-frequency property.
Consequently, if the bandwidth of g(t) is B Hz, then the bandwidth of
g (t) is 2B Hz, and the bandwidth of g n (t) is nB Hz. We mentioned this
2

property in 2.26.
5.4. BW Inefficiency in DSB-SC: Recall that for real-valued baseband signal
m(t), the conjugate symmetry property from 2.16 says that

M (−f ) = (M (f ))∗ .
7
This property states that the width of x ∗ y is the sum of the widths of x and y.
8
The width property of convolution does not hold in some pathological cases. See [4, p 98].

33
The DSB spectrum has two sidebands: the upper sideband (USB) and the
lower sideband (LSB), both containing complete information about the base-
band signal m(t). As a result, DSB signals occupy twice the bandwidth
required for the baseband. To improve the spectral efficiency of amplitude
modulation, there exist two basic schemes to either utilize or remove the
spectral redundancy:
(a) Single-sideband (SSB) modulation, which removes either the LSB or
the USB so that for one message signal m(t), there is only a bandwidth
of B Hz.
(b) Quadrature amplitude modulation (QAM), which utilizes spectral re-
dundancy by sending two messages over the same bandwidth of 2B
Hz.
We will only discussed QAM here. SSB discussion can be found in [3, Sec
4.4], [10, Section 3.1.3] and [4, Section 4.5].
Definition 5.5. In quadrature amplitude modulation (QAM ) or quadra-
ture multiplexing , two baseband real-valued signals m1 (t) and m2 (t) are
transmitted simultaneously via the corresponding QAM signal:
√ √
xQAM (t) = m1 (t) 2 cos (ωc t) + m2 (t) 2 sin (ωc t) .

• QAM operates by transmitting two DSB signals via carriers of the same
frequency but in phase quadrature.
• QAM can be exactly generated without requiring sharp cutoff bandpass
filters.

34
 • Both modulated signals simultaneously occupy the same frequency
band.
• The upper channel is also known as the in-phase (I ) channel and the
lower channel is the quadrature (Q) channel.
5.6. Demodulation: The two baseband signals can be separated at the
receiver by synchronous detection:
n √ o
LPF xQAM (t) 2 cos (ωc t) = m1 (t)
n √ o
LPF xQAM (t) 2 sin (ωc t) = m2 (t)

• m1 (t) and m2 (t) can be separately demodulated.

5.7. Sinusoidal form:
√
xQAM (t) = 2E(t) cos(2πfc t + θ(t)),

where q
E(t) = m21 (t) + m22 (t)
 
m2 (t)
θ(t) = − tan−1
m1 (t)
5.8. Complex form:
√
2Re (m(t)) ej2πfc t

xQAM (t) =

where m(t) = m1 (t) − jm2 (t).

35
 • If we use − sin(ωc t) instead of sin(ωc t),
√ √
xQAM (t) = m1 (t) 2 cos (ωc t) − m2 (t) 2 sin (ωc t)

and
m(t) = m1 (t) + jm2 (t).

• We refer to m(t) as the complex envelope (or complex baseband

signal ) and the signals m1 (t) and m2 (t) are known as the in-phase
and quadrature(-phase) components of xQAM (t).
• The term “quadrature component” refers to the fact that it is in phase
quadrature (π/2 out of phase) with respect to the in-phase component.
• Key equation:
 
 
√ j2πf t o √ −j2πf t 

 n 
LPF Re m (t) × 2e c × 2e c
= m (t) .

 | {z } 

 
x(t)

5.9. Three equivalent ways of saying exactly the same thing:

(a) the complex-valued envelope m(t) complex-modulates the complex car-
rier ej2πfc t ,
(b) the real-valued amplitude E(t) and phase θ(t) real-modulate the am-
plitude and phase of the real carrier cos(ωc t),
(c) the in-phase signal m1 (t) and quadrature signal m2 (t) real-modulate the
real in-phase carrier cos(ωc t) and the real quadrature carrier sin(ωc t).
5.10. References: [3, p 164–166], [10, Sect. 2.9.4], [4, Sect. 4.4], and [7,
Sect. 1.4.1]
5.11. Question: In engineering and applied science, measured signals are
real. Why should real measurable effects be represented by complex signals?
Answer: One complex signal (or channel) can carry information about
two real signals (or two real channels), and the algebra and geometry of
analyzing these two real signals as if they were one complex signal brings
economies and insights that would not otherwise emerge.

36
6 Amplitude modulation: AM
6.1. The analysis of DSB-SC in the earlier sections illustrates that the
spectrum of a DSB signal does not contain a discrete spectral component
at the carrier frequency unless m(t) has a DC component. This is why we
referred to it as a suppressed carrier system.
6.2. DSB-SC amplitude modulation is easy to understand and to analyze
in both time and frequency domains. However, analytical simplicity is not
always accompanied by an equivalent simplicity in practical implementation.
Problem: The (coherent) demodulation of DSB-SC signal requires the
receiver to possess a carrier signal that is synchronized with the incoming
carrier. This requirement is not easy to achieve in practice because the
modulated signal may have traveled hundreds of miles and could even suffer
from some unknown frequency shift.
6.3. If a carrier component is transmitted along with the DSB signal,
demodulation can be simplified.
(a) The received carrier component can be extracted using a narrowband
bandpass filter and can be used as the demodulation carrier. (There is
no need to generate a carrier at the receiver.)

(b) If the carrier amplitude is sufficiently large, the need for generating a
demodulation carrier can be completely avoided.
• This will be the focus of this section.
Definition 6.4. For AM, the transmitted signal is typically defined as

xAM (t) = (A + m (t)) cos (2πfc t) = A cos (2πfc t) + m (t) cos (2πfc t)
| {z } | {z }
carrier sidebands

37
6.5. Trade-off:
(a) Disadvantage:
• Higher power and hence higher cost required at the transmitter
• The carrier component is wasted power as far as information trans-
fer is concerned.
• This fact can completely preclude the use of AM in power-limited
applications.
(b) Advantage:
• Coherent reference is not needed for demodulation.
• Demodulator becomes simple and inexpensive.
• For broadcast system such as commercial radio (with a huge num-
ber of receivers for each transmitter,
◦ any cost saving at the receiver is multiplied by the number of
receiver units.
◦ it is more economical to have one expensive high-power trans-
mitter and simpler, less expensive receivers.
(c) Conclusion: Broadcasting systems tend to favor the trade-off by mi-
grating cost from the (many) receivers to the (fewer) transmitters.
6.6. Spectrum of xAM :

• Basically the same as that of DSB-SC except for the two additional
impulses at ±fc .
Definition 6.7. Consider a signal A(t) cos(2πfc t). If A(t) varies slowly in
comparison with the sinusoidal carrier cos(2πfc t), then the envelope E(t)
of A(t) cos(2πfc t) is |A(t)|.

38
 6.8. Envelope of AM signal : See Figure 9. For AM signal, A(t) =
A + m(t).
(a) If ∀t, A(t) > 0, then E(t) = A(t) = A + m(t)
• The envelope has the same shape as m(t).
• We can detect the desired signal m(t) by detecting the envelope
(envelope detection).
(b) If ∃t, A(t) < 0, then E(t) 6= A(t).
• The envelope shape differs from the shape of m(t) because the
192
negative part of A + m(t) is rectified.
AMPLITUDE MODULATIONS AND DEMODULATIONS

Figure 4.8 m(1)

AM signal and
its enve lope. ~np

A+ m(l) >0 far all 1

A + m(1) ':{> 0 far all 1

tA
t
1~
}L\-------
t (~ 1--
(b)

En velope
Envelope
~A+ m(l)
lA + m(t)l

I (d) (e)
I

Figure 9: AM signal and its envelope [5, Fig 4.8]

and m(t) cannot be recovered from the envelope. Consequently, demodul ation of rpAM (t) in
Fig. 4.8d amounts to sim ple envelope detection. Thus, the condition for envelope detection
of an AM signal is
39
A + m (t) ::: 0 for all t (4.9a)

If m(t) ::: 0 for all/ , then A = 0 already sati sfies condition (4 .9a) . In thi s case there is no need
6.9. Summary of AM Concept:
• The carrier term A cos(2πfc t) is added.
• The size of A affects the time domain envelope of the modulated signal.
• A should be large enough to ensure that A+m(t) is always nonnegative.
◦ If ∀t, m(t) ≥ 0, then there is no need to add any carrier. The
DSB-SC signal can be detected by envelope detection.
6.10. Demodulation of AM Signals via rectifier detector: The receiver
will first recover A + m(t) and then remove A. Note that, conceptually, the
received signal is the same as DSB-SC signal except that the m(t) in the
DSB-SC signal is replaced by A(t) = A + m(t). We will also assume that A
is large enough so that A(t) ≥ 0.
Recall the key equation of switching demodulator (38):
1
LPF{A(t) cos(2πfc t) × 1[cos(2πfc t) ≥ 0]} = A(t) (39)
π
We noted before that this technique requires the switching to be in sync
with the incoming cosine.
When ∀t, A(t) ≥ 0, we can replace the switching demodulator by the rec-
tifier demodulator/detector . In which case, we suppress the negative
part of m(t) cos(ωc t) using a diode (half-wave rectifier). This is mathemat-
ically equivalent to switching demodulator in (38) and (39).

40
6 AMPLITUDE MODULATIONS AND DEMODULATIONS

gure 4.10
[a+ m(t)] cos wet VR(t) /[A + m(t)]
ctifier detector
AM. -_f I
" rr [A + 111(1)] I
' -;-[A + m(1)]
~
-·
' /

Low-pass
[A + m(l)] cos wet filter

Figure 10: Rectifier detector for AM [5, Fig. 4.10].

signal is multiplied by w(t). Hence, the half-wave rectified output vR(t) is

• It={[A+
VR(t) is in m(t)]
effectCOSsynchronous
Wet) w(t) detection performed without using a(4.12)
local
carrier [4, p 167].
=[A+ m(t)] cos Wet [ ~ + ~ (cos (Vet- ~cos 3wet + ~cos Swet- · · ·)] (4.13)
• This method needs A(t) ≥ 0 so that the sign of A(t) cos(ωc t) will be
l
the same m(t)]
= -[A+
][
as the signterms
+other of cos(ω c t). frequencies
of higher (4.14)

A
When• The
vR(t) dc term toπ amay
is applied be blocked
low-pass by aBcapacitor
filter of cutoff to isgive
Hz, the output [A+the desired
m(t)]jn, and output
all the
m(t)/π.
other term s of frequencies higher than B Hz are suppressed. The de term Ajn may be blocked
by a capac itor (Fig. 4.10) to give the desired output m(t) j n. The outp ut can be doubled by
6.11.
using Demodulation
a full-wave rectifi er. of AM signal via envelope detector :
It is interesting to note that because of the multip lication with ll '(l), rectifier detection is in
• Design criterion
effect synchronous detectionofperformed
RC: without using a local carrier. The high carrier content
in AM ensures that its zero crossings are periodic1 and the informatio n abo ut frequency and
phase of the carrier at the transmitter is built
2πB in to the 
AM2πf c . itself.
signal
RC
Envelope Detector: fn an enve lope detector, the o utput of the detector follows the
• The envelope detector output is A + m(t) with a ripple of frequency f .
envelope of the modulated signal. The simpl e circuit show n in Fig. 4. lla functions as an c
envelope detector. On the positive cycle of the input signa l, the input grows and may exceed
• The dc term can be blocked out by a capacitor or a simple RC high-pass
the charged vo ltage on the capacity vc(t), turning on the diode and allow ing the capac itor C
filter.
to charge up to the peak voltage of the input signal cycle. As the input signal fall s below this
peak value, it falls quickly below the capacitor voltage (which is very nearly the peak voltage),
• The ripple may be reduced further by another (low-pass) RC filter.
thus caus ing the diode to open. The capacitor no w di scharges through the resi stor R at a slow
rate (with a time constant RC). During the next positive cycle, the same drama repeats . As the
6.12.
input References:
signal [3, capacitor
rises above the p 198–199], [5,the
voltage, Section 4.3] and
diode conducts [10,The
again. Section
capacitor3.1.2].
again
charges to the peak value of this (new) cycle. The capacitor discharges slowly during the cutoff
period.
During each positive cycle, the capacitor41charges up to the peak voltage of the input
signal and then decays slowly until the next positive cycle, as shown in Fig. 4 . ll b. Thus, the
output voltage vc(t), close ly follows the (rising) envelope of the input AM signal. Equally
important, the slow capacity discharge via the resistor R a ll ows the capacity vo ltage to follow
 4 .4 Bandw idth-Efficient Amplitude Modulati ons 197

Figure 4.11
Envelope
detector for AM.

AM signal c

(a)
Envelope de tector output

RC too large \
····· K' f<K~
-- ~. . Envelop~.--· ... ·· · ' ( KI"' I""
~-~ i" " !'--
, .,.
W'~
,. ·<
~·· · ·~"

....
... -·· '

..··· ·...
·· .. ..
(b) ······

Figure 11:
a declining envelope. Envelope
Capacitor detector
d ischarge betwee nfor AM
positi ve [5, Fig.
peaks 4.11].
ca uses a rippl e signal of
freque ncy We in the output. Thi s rip ple can be reduced by choosing a larger time constant
RC so that the capac ito r disc harges ve ry littl e between the positive peaks (RC » I /eve) . If
RC were made too large, however, it wo uld be imposs ible for the capac itor voltage to follow
a fast declining e nvelope (Fig. 4.11 b). Because the max imum rate of AM envelope dec line
is do minated by the ba ndw idth B of the message sig nal m (r ) , the des ign criterion of R C
should be

I
I /eve « RC < I / (2Jr8) or 2Jr8 < -
RC
« (t!c

The envelope detector output is vc( t ) = A+ m(r) w ith a rippl e o f frequency W e . The de term
A ca n be bl ocked oul by a capacitor or a simpl e RC hig h-pass filte r. The rippl e may be reduced
further by another (low-pass) R C filter.

42
4.4 BANDWIDTH-EFFICIENT AMPLITUDE
MODULATIONS
As seen from Fig . 4.12, the DSB spectrum (including suppressed carrier and AM) has two
sidebands: the upper sideband (USB) and the lower sideband (LSB~both containing complete
 Sirindhorn International Institute of Technology
Thammasat University
School of Information, Computer and Communication Technology

ECS332 2012/1 Part II.3 Dr.Prapun

7 Angle Modulation: FM and PM
7.1. Recall that a sinusoidal carrier signal

A cos(2πfc t + φ)

has three basic parameters: amplitude, frequency, and phase. Varying these
parameters in proportion to the baseband signal results in amplitude mod-
ulation (AM), frequency modulation (FM), and phase modulation (PM),
respectively.
7.2. As usual, we will again assume that the baseband signal m(t) is band-
limited to B; that is, |M (f )| = 0 for |f | > B.
In this section, we will also assume that

|m(t)| ≤ mp .

In other words, m(t) is bounded between −mp and mp .

43
Definition 7.3. The main characteristic9 of frequency modulation is
that the carrier frequency f (t) would be varied with time so that

f (t) = fc + km(t), (40)

where k is an arbitrary constant.

• The arbitrary constant k is sometimes denoted by kf to distinguish it
from a similar constant in PM.
7.4. FM: A Magical Technique?
In the 1920s, the idea of frequency modulation (FM) was quite magical.
FM was naively proposed very early as a method to conserve the radio
spectrum. The naive argument was presented as followed:
• If m(t) is bounded between −mp and mp , then the maximum and mini-
mum values of the (instantaneous) carrier frequency would be fc + kmp
and fc − kmp , respectively. (Think of this as a delta function shifting
to various location between fc + kmp and fc − kmp in the frequency
domain.)

• Hence, the spectral components would remain within this band with a
bandwidth 2kmp centered at fc .
• Conclusion: By using an arbitrarily small k, we could make the infor-
mation bandwidth arbitrarily small (much smaller than the bandwidth
of m(t).
In 1922, Carson argued that this is an ill-considered plan. We will illustrate
his reasoning later. In fact, experimental results shows that

As a result of his observation, FM temporarily fell out of favor.

9
Treat this as a practical definition. The more rigorous definition will be provided in 7.11.

44
 7.5. Armstrong (1936) reawakened interest in FM when he realized it had
a much different property that was desirable. When the kf is large, the
inverse mapping from the modulated waveform xF M (t) back to the signal
m(t) is much less sensitive to additive noise in the received signal than is
the case for amplitude modulation. FM then came to be preferred to AM
because of its higher fidelity. [1, p 5-6]

7.1 Instantaneous Frequency

To understand more about FM, we will first need to know what it actually
means to vary the frequency of a sinusoid.
Definition 7.6. The generalized sinusoidal signal is a signal of the form

x(t) = A cos (θ(t)) (41)

where θ(t) is called the generalized angle.
• The generalized angle for conventional sinusoid is ωc t + θ0 .
7.7. Suppose we want the frequency fc of a carrier A cos(2πfc t) to vary
with time as in (40). It is tempting to consider the signal

A cos(2πf (t)t),

where f (t) is the desired frequency at time t.

Example 7.8. See Slides. Consider the generalized sinusoid with f (t) = t2 .
Definition 7.9. For generalized sinusoid A cos(θ(t)), the instantaneous
frequency 10 at time t is given by
1 d
f (t) = θ(t). (42)
2π dt
7.10. Equation (42) implies
Z t Z t
θ(t) = 2π f (τ )dτ = θ(t0 ) + 2π f (τ )dτ. (43)
−∞ t0
10
Although f (t) is measured in hertz, it should not be equated with spectral frequency. Spectral frequency
f is the independent variable of the frequency domain, whereas instantaneous frequency f (t) is a time-
dependent property of waveforms with exponential modulation.

45
Definition 7.11. Frequency modulation (FM ):
 
Zt
xFM (t) = A cos 2πfc t + φ + 2πkf m (τ )dτ  .
−∞

The instantaneous frequency is given by

f (t) = fc + kf m (t) .
Definition 7.12. Phase modulation (PM ):
xPM (t) = A cos (2πfc t + φ + kp m (t))
The instantaneous frequency is given by

7.13. Generalized angle modulation (or exponential modulation):

x(t) = A cos (ωc t + θ0 + (m ∗ h)(t))
where h is causal.
(a) Frequency modulation (FM ): h(t) = 2πkf 1[1 ≥ 0]
(b) Phase modulation (PM ): h(t) = kp δ(t).

Frequency modulator
t

m (t ) ∫ m (τ )dτ Phase xFM ( t )

∫ −∞
Modulator

m (t ) d m′ ( t ) Frequency xPM ( t )
dt Modulator

Phase modulator

Figure 12: With the help of integrating and differentiating networks, a phase modulator
can produce frequency modulation and vice versa [4, Fig 5.2].

46
A Trig Identities
All of the trigonometric functions of an angle θ can be constructed geo-
metrically in terms of a unit circle centered at origin as shown in Figure
13.
F co
excsc t
cvs A

csc tan
sin sin

θ C
cos versin D exsec E

sec
1
Unit circle
B

Figure 13: Trigonometric functions on a unit circle.

A.1. Cosine function

(a) Is an even function: cos(−x) = cos(x).
(b) cos x − π2 = sin(x).

(c) Sum formula:

cos(x ± y) = cos x cos y ∓ sin x sin y. (44)

(d) Product-to-Sum Formula:

1
cos(x) cos(y) = (cos(x + y) + cos(x − y)) .
2

n−1
 2
1 n
P

cos ((n − 2k) x), odd n ≥ 1


2n−1 k


n k=0
(e) cos x = n
!
2 −1
1 n n
 P




 2n 2 k cos ((n − 2k) x) + n , even n ≥ 2
 2
k=0

47
(f) Any two real numbers a, b can be expressed in terms of cosine and sine
with the same amplitude and phase:

(a, b) = (A cos(φ), A sin(φ)) , (45)

√
where A = a2 + b2 and φ = tan−1 ab . This is simply the polar-
coordinates from of the point (a, b) on Cartesian coordinates.
A.2. Properties of eix

(a) Euler’s formula: eix = cos x + i sin x. Hence,

1 jA
cos (A) = Re ejA = e + e−jA


2  
1 1 jA
sin (A) = Im ejA = Re −jejA = Re + ejA = e − e−jA .




j 2j

• We can use cos x = 12 eix + e−ix and sin x = 2i1 eix − e−ix to

derive many trigonometric identities.

In fact, we can combine linear combination of cosine and sine of the
same argument into a single cosine by
 
p
−1 B
A cos ω0 t + B sin ω0 t = A2 + B 2 cos ω0 t − tan .
A
To see this, note that
A cos ω0 t + B sin ω0 t = Re Aejω0 t + Re −jBejω0 t = Re (A − jB) ejω0 t




p 
2 2 −j tan−1 B jω0 t
= Re A +B e A e .

Another way to see this is to reexpress the two real numbers A, B using
(45) and then use (44).
(b) ejx is periodic with period 2π.
−1 y
x2 + y 2 ej tan ( x ) =
p
(c) Any complex number z = x+jy can be expressed as z =
|z|ejφ .
• z t = |z|t ejφt .
(d) More relations involving sin and cos.

48
 A+B A−B


• ejAt + ejBt = 2ej 2 t cos 2 .
A+B A−B


• ejAt − ejBt = 2jej 2 t sin 2
ejAt −ejBt (A+B)−(C+D) sin A−B( )
• ejCt −ejDt = ej 2 t 2
.
(
sin C−D
2 )

49
References
[1] Richard E. Blahut. Modem Theory: An Introduction to Telecommuni-
cations. Cambridge University Press, 1 edition, December 2009. 7.5
[2] Albert Boggess and Francis J. Narcowich. First Course in Wavelets
with Fourier Analysis. Prentice Hall, 1 edition, 2001. 2.27, 2.28
[3] A. Bruce Carlson and Paul B. Crilly. Communication Systems: An In-
troduction to Signals and Noise in Electrical Communication. McGraw-
Hill, 5th international edition edition, 2010. 1, 2, 3, 4, 5.4, 5.10, 6.12
[4] B. P. Lathi. Modern Digital and Analog Communication Systems. Ox-
ford University Press, 1998. 2.2, 2, 3, 3.10, 4.5, 4.16, 8, 4.18, 8, 5.4,
5.10, 6.10, 12
[5] B. P. Lathi and Zhi Ding. Modern Digital and Analog Communication
Systems. Oxford University Press, 2009. 9, 10, 6.12, 11
[6] C. Britton Rorabaugh. Communications Formulas and Algorithms: For
System Analysis and Design. Mcgraw-Hill, 1990.
[7] Peter J. Schreier and Louis L. Scharf. Statistical Signal Processing of
Complex-Valued Data: The Theory of Improper and Noncircular Sig-
nals. Cambridge University Press, 2010. 6, 5.10
[8] Claude E. Shannon. A mathematical theory of communication. Bell
Syst. Tech. J., 27(3):379–423, July 1948. Continued 27(4):623-656, Oc-
tober 1948. 1.1, 1.2
[9] Elias M. Stein and Rami Shakarchi. Fourier Analysis: An Introduction.
Princeton University Press, March 2003. 2.28
[10] Rodger E. Ziemer and William H. Tranter. Principles of Communica-
tions. John Wiley & Sons Ltd, 2010. 2.9, 3.11, 5.4, 5.10, 6.12

50