Chapter 3: Characterization of

Communication Signals and Systems

Graduate Program
School of Electrical and Computer Engineering
Goals of the Chapter
• Signals can be categorized in a number of ways
• Will cover characterization of signals and systems
encountered in digital communication systems
• Focus points
• Characterization of bandpass signals and systems
• Vector space representation of signals
• Representation of digitally modulated signals

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 2

Overview
• Signals and systems
• Stationary stochastic process
• Signal space

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 3

Communication Signals & Systems Characterization
• Binary bits from the source encoder mapped into signal
waveforms, mostly after channel encoding
• In channel encoding redundancy is added in a controlled manner
for error correction at the receiver
• Example: Binary modulation: 0 → s1(t) and 1→ s2(t)
• b bits at a time mapped using M= 2b waveforms
si(t), i = 0,1,2,…M-1
• I.e., one waveform for each of the 2b possible bit sequences
• M-ary modulation for M > 2
• Question: What should be the characteristics of these
waveforms and how do we describe and use them?
• Various forms of digitally modulated signals will be introduced
along with their spectral and other characteristics

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 4

Bandpass Signals and Systems
• Channels have limited bandwidth centered about the
carrier (DSB) or adjacent to the carrier (SSB)
• Narrowband Definition: BW<< fc, carrier frequency
• Reduce all bandpass signals and channels to equivalent
lowpass signals and channels
• Without any loss of generality, it makes the analysis
independent of the carrier frequency
• We consider that s(t) has frequency content around a
narrowband in the vicinity of the frequency, fc

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 5

Bandpass Signals and Systems
• Construct an analytic signal that contains only the positive
frequencies
∧
S + (t ) = S (t ) + j S (t ) OR S + ( f ) = 2U ( f ) S ( f )
∞
∧ 1 s (τ )
Where S (t ) = ∫
π −∞ t − τ
dτ , Hilbert transform of s (τ )

• The analytic signal S+(t) is a bandpass signal

• Equivalent lowpass representation can be obtained by
frequency translation
S l (t ) = S + (t ) e − j 2πf c t 
=  S (t ) + j S (t ) e − j 2πf c t
∧
OR
 
∧
S (t ) + j S (t ) = S l (t ) e j 2πf c t

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 6

Bandpass Signals and Systems …
• Since Sl(t) is in general complex, it may be expressed as
Sl (t) = x(t)+ jy(t) then
∧
S(t)+ j S (t) = ( x(t)+ jy(t))(cos 2 πf c t + jsin 2 πf c t )
= ( x(t) cos 2 πf c t − y(t) sin 2 πf c t ) +
j ( x(t) sin 2 πf c t + y(t) cos 2 πf c t )

• Equating the real and imaginary parts

S (t ) = x(t ) cos 2πf c t − y (t ) sin 2πf c t
∧
(1)
S (t ) = x(t ) sin 2πf c t + y (t ) cos 2πf c t

• S(t) is the desired form of the bandpass signal

• x(t) and y(t) are amplitude modulations impressed on carriers
cos 2πfct and sin2πfct, which are in phase quadrature
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 7
Bandpass Signals and Systems …
• Alternatively,

{ } [
S(t) = Re [(x(t)+ jy(t )]e j 2 πf ct = Re Sl (t) e j 2 πf ct ] (2)

• A third possible form of representation of S(t) can be

S l (t) = a(t) e jθ (t) Where

−1 y(t) (3)
a(t) = (x (t) + y (t))
2 2
and θ(t) = tan
x(t)
( )
S(t) = Re Sl (t) e j 2 πf ct = a(t) cos(2 πf c t + θ(t))

• a(t) and θ(t) are the envelope and the phase angle of S(t),
respectively

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 8

Bandpass Signals and Systems …
• The signal can also be expressed in frequency domain
through its Fourier transform
∞ ∞
S ( f ) = ∫ S (t ) e − j 2πft dt = ∫ Re S (
l (t ) e j 2πf c t
)
e − j 2πft
dt
−∞ −∞

• And using Re( Z ) = 1 (Z + Z * )

2

∞
1
[
• We get S ( f ) = ∫ Sl (t ) e j 2πf ct + Sl* (t ) e − j 2πf ct e − j 2πft dt
2 −∞
]
=
1
2
[
Sl ( f − f c ) + Sl* (− f − f c ) ; ]
• Where Sl(f) is the transform of Sl(t)
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 9
Bandpass Signals and Systems …
• The energy in the signal S(t) is given by
∞ ∞
ε = ∫ S 2 (t) dt = ∫ {Re[S (t) e ]} dt
i 2 πf c t 2
l
−∞ −∞
∞ ∞
1 1
= ∫ Sl (t) dt + ∫ Sl (t) 2 cos(4 πft + 2 θ(t) ) dt
2

2 −∞ 2 −∞

• Since in the second term │Sl(t)│ varies very slowly its

contribution may be neglected and the energy of the signal
given by the first element of the sum only
∞
1
ε ≈ ∫ Sl (t ) 2 dt
2 −∞

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 10

Bandpass Signals and Systems …

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 11

Linear Bandpass System
• Linear bandpass systems are characterized by the
impulse response h(t) or by the frequency response H(f),
which is the Fourier transform of h(t)
• Note that for real h(t), H*(-f)= H(f)
• Define:
H ( f ) f >0
H l ( f − fc ) = 
0 f <0

• Then:
0 f >0
H (− f − f c ) = 
*

f <0
l
H ( f )
∴ H ( f ) = H l ( f − f c ) + H l* (− f − f c )

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 12

Linear Bandpass System
• Or in time domain using the inverse transform

h(t) = hl (t) e j 2 πf ct + hl* (t) e − j 2 πf ct

[
= 2 Re hl (t) e j 2 πf c t
]
• Where hl(t) and Hl(f) are Fourier transform pairs and are in
general complex valued functions that characterize the
equivalent lowpass system

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 13

Response of Bandpass System to Bandpass Signal
• Assume an input signal S(t) is a narrowband bandpass
(BP) signal and the system is also narrowband BP

S(t) r(t)
h(t) or hl(t)
Sl (t) rl (t)
j 2 πf c t
r(t) = Re (rl (t)e )
∞

∫
• Where r(t) = S(τ ) h(t − τ) dt or
−∞
R(f) = S(f) H(f)

1
R( f ) = [ Sl ( f − f c ) − S*l ( −f − f c )] [ H l ( f − f c ) − H*l ( −f − f c )]
2
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 14
Response of Bandpass System to Bandpass Signal
• For narrowband signal and narrowband impulse response

Sl ( f − f c ) ≈ 0 and H l ( f − f ) = 0 for f < 0

Thus Sl ( f − f c ) H l* (− f − f c ) = Sl* (− f − f ) H l ( f − f ) = 0

1
R ( f ) ≈ [ sl ( f − f c ) H l ( f − f c ) + Sl* ( f − f c ) H l* (− f − f c )]
2
1
= [ Rl ( f − f c ) + Rl* (− f − f c )]
2
• Where
Rl ( f ) = Sl ( f ) H l ( f )
• is output spectrum of the LPF system excited by LP signal
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 15
Overview
• Signals and systems
• Stationary stochastic process
• Signal space

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 16

Bandpass Stationary Stochastic Processes
• Suppose n(t) is a sample function of a wide sense
stationary (WSS) stochastic process with zero mean and
power spectral density Φnn(f)
• Φnn(f) is assumed zero outside an interval Δf centered around ± fc
• n(t) is narrowband process if Δf << fc
• n(t) may be represented by any of the following three forms
n(t) = a(t) cos(2 πf c t + θ(t))
= x(t) cos 2 πf c t − y(t) sin 2 πf c t
(
= Re z(t) e − j 2 πf ct )
• Where
z(t) = x(t) + jy(t)
−1 y (t )
θ (t ) = tan and a(t) = x 2 (t ) + y 2 (t )
x(t )

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 17

Bandpass Stationary Stochastic Processes …
• Since n(t) is zero-mean, x(t) and y(t) are also zero mean
• Furthermore, from the stationarity of n(t) follows
φ xx (τ ) = φ yy (τ ); φ xy (τ ) = −φ yx (τ ) and
φnn (τ ) = φ xx (τ ) cos 2πf cτ − φ yx (τ ) sin 2πf cτ

• Which is identical in form with the expression for n(t)

• The autocorrelation of the equivalent lowpass process
z (t ) = x(t ) + jy (t ) is
1
{ }
φ zz (τ ) = E z ∗ (t ) z (t + τ ) = φ xx (τ ) + jφ yx (τ )
2
• Which is the autocorrelation of the complex evelope
• Finally, φnn (τ ) = Re(φ zz (τ )e j 2πf cτ )
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 18
Bandpass Stationary Stochastic Processes …
• Thus Φnn(τ) of the bandpass stochastic process is
uniquely determined from the autocorrelation function
ΦZZ(τ) of the equivalent lowpass process z(t) and carrier
frequency fc
• Note that
∞

∫ [Re(φ )]e − j 2 πf τ
j 2πf cτ
φnn ( f ) = zz (τ )e dτ
−∞

= [φ zz ( f − f c ) + φ zz (− f − f c )]
1
2
• Where ΦZZ(f) is the power spectrum of the lowpass
process z(t)
• Since ΦZZ(τ) = ΦZZ*(-τ), it follows that ΦZZ(f) is real valued
function of frequency
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 19
Bandpass Stationary Stochastic Processes …
• φ xy (τ ) is an odd function of τ and φ xy (0) = 0 and hence x(t)
and y(t) are uncorrelated for τ = 0
• If n(t) is Gaussian x(t) and y(t) are jointly Gaussian and for
τ = 0 are independent

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 20

Representation of White Noise
• White noise is wideband and cannot be represented in
terms of quadrature components
• If the noise is assumed to have passed through an ideal
bandpass filter, the output can be represented by
quadrature components
Φnn(f)

1/2N0

B B
White BP white
noise noise
-fc fc

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 21

Representation of White Noise …
• The equivalent lowpass noise z(t) has a power spectral
density:
 1
 N0 f ≤ B
φ zz ( f ) =  2
1
0 f > B
 2
• And autocorrelation function is:
sin πBτ
φ zz (τ ) = N 0 which → N 0δ (τ ) as B → ∞
πBτ
φ yx (τ ) = 0 for all τ and φ zz (τ ) = φ xx (τ ) = φ yy (τ )

• i.e the quadrature components x(t) and y(t) are

uncorrelated for all time shifts τ and the autocorrelation of
z(t), x(t) and y(t) are all equal
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 22
Overview
• Signals and systems
• Stationary stochastic process
• Signal space

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 23

Signal Space Concepts
• Analogous to space vectors, we represent a family of
signals such as X = { x0 (t ), x1 (t ),..........xM −1} by a signal
space over a given time interval
• A signal space is defined by its orthonormal basis
{f 0 (t ), f1 (t ),................. f N −1 }
• The inner product of two signals is defined as
b
x1 (t) , x2 (t) = ∫ x1 (t) x*2 (t) dt
a

• The norm of a signal is defined as

1
  b 2

 ∫ x(t) dt 
x(t) = 
2

a 
Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 24
Signal Space Concepts …
• A set of signals are orthonormal iff they are orthogonal and
their norms are each unity
• The set { f 0 (t ), f1 (t ),.............. f N −1 } is an orthonormal basis of
the signal X iff
N −1
∀ x j (t ) ∈ X , ∃ ai ∈ ℜ x j (t ) = ∑ ai f i (t )
i =0

• Let s(t) be a deterministic, real valued signal with finite

energy ∞
Es = ∫
−∞
s 2 (t ) dt

• And let the set { f n (t )}, n = 0, 1, 2,..............., N − 1 be an orthonormal

set of signals or waveforms

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 25

Signal Space Concepts …
• We can approximate s(t) by
∧ N −1
S (t) = ∑s
n =0
n f n (t), where
∞
sn = ∫ s(t) f
−∞
n (t), n = 1, 2,............ N − 1

• The set of waveforms { f n (t )}, n = 0,1, 2,...............N − 1 is said to

be complete iff the error energy is zero, i.e.,
∞ 2
N −1
  ∧
ε e = ∫  s(t ) − s (t )  dt = E s − ∑ s n2 = 0
− ∞  n =0

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 26

Signal Space Concepts …
• Consider the signal x(t) given by
1 − 0.5 ≤ t ≤ 0.5
x(t) = 
0 elsewhere
• We approximate x(t) by
K
x(t) = ∑ xn cos nπ t −1≤ t ≤ 1
n =0

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 27

Signal Space Concepts …

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 28

Signal Space Concepts …
• In general, assume that we have a set of waveforms
(signals)
{s0 (t ), s1 (t ),..................s M −1 (t )}
• And we wish to construct a set of orthonormal waveforms
{f 0 (t ), f 1 (t ),.............. f N −1 }
from the original signal set
• This can be done either:
• Formally by using the Gram-Schmidt orthogonalization
procedure (READ Page 163 of the text); or
• In simple cases, by inspection

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 29

Signal Space Concepts …
• Example: Find an orthonormal basis for the following set
of waveforms and determine the coordinates of each
waveform in the signal space defined by the orthonormal
basis functions

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 30

Signal Space Concepts …
• By inspection the following orthonormal function can be
the basis for the representation of the three waveforms
shown above

1 1
s1 (t ) = f1 (t ); s2 (t ) = f1 (t ) − f 2 (t ) + f 3 (t )
2 2
1 1
s3 (t ) = f1 (t ) + f 2 (t ) + f 3 (t )
2 2
In vector form
 1   1 1 
S s = [1,0 ,0];
1
S 2 = 1,− , ; S 3 = 1, ,
 2 
2  2 2 

Sem. I, 2012/13 Digital Communications – Chapter 3: Communication Signals & Systems 31