Signals and Systems

Lecture 2-5
25
BITS Pilani
K K Birla Goa Campus
p Nitin Sharma
Outline

• Size of Signal…
• Classification of Signals
• Examples of Signals
• S
Some useful
f l Signal
Si lOOperations
ti
• Some useful Signals
• Unit Impulse Function
• Unit Step Function
• The Exponential Function

BITS Pilani, K K Birla Goa Campus

 Size of a Signal(1)
• The size of any entity is a number that indicates the largeness
or strength of that entity.
entity
• Generally speaking, the signal amplitude varies with time.
• How can a signal that exists over a certain time interval with
varying amplitude be measured by one number that will
indicate the signal size or signal strength?
• Such
S h a measure mustt consider id nott only
l the
th signal
i l amplitude,
lit d
but also its duration.
• So,
So area under a signal x(t) gives possible measure of its size
• To account for positive and negative areas cancellation, we
find area under x2(t), which is always positive

BITS Pilani, K K Birla Goa Campus

Size of a Signal(2)

• Measured by signal energy Ex:

• Generalize
G li ffor a complex
l valued
l d signal
i l tto:

• A signal with finite energy is an energy signal.

• Energy
gy must be finite,, which means ((Power is zero))

• Else integral will not converge.

BITS Pilani, K K Birla Goa Campus
 Size of a Signal(3)
• A signal with infinite energy is a power signal.
• If amplitude of x(t) does not → 0 when t → ∞ ∞, we find time average
of the energy [mean-squared value of x(t)]) denoted by Px :

• Again, generalize for a complex valued signal to:

• The mean of an entity averaged over a large time interval

approaching infinity exists if the entity is either periodic or has a
statistical regularity.
• If such a condition is not satisfied, the average may not exist.
BITS Pilani, K K Birla Goa Campus
 Size of a Signal(4)
• Signal with finite energy (zero power)

• Signal with finite power (infinite energy)

BITS Pilani, K K Birla Goa Campus

Signals Classification

Signals may be classified into:

- Discrete-time and continuous-time signals
- Analogue and digital signals
- Periodic
P i di and d aperiodic
i di signals
i l
- Deterministic and Random/Probabilistic signals
- Energy and power signals
- Causal and Non-causal signal
- Even and Odd Signals

BITS Pilani, K K Birla Goa Campus

Continuous-Time & Discrete-
Time
Modern systems generally…
– get a continuous-time
continuous time signal from a sensor
– a cont.-time system modifies the signal
– an “analog-to-digital converter” (ADC or A-to-D) sample the signal
– to create a discrete-time signal
g … a “stream of numbers”
– A discrete-time system to do the processing
– convert back to C-T signal with a “Digital-to-Analog Converter” (DAC)

BITS Pilani, K K Birla Goa Campus

Sampling: Time-Domain Plot

BITS Pilani, K K Birla Goa Campus

Quantization: Illustration

2m p
v 
L

Quantizer Input Samples x

Quantizer Output Samples x q

BITS Pilani, K K Birla Goa Campus

 Continuous vs Discrete
(Qualifies nature of x axis)
• A signal that is specified for every value of time t is a
continuous time signal
continuous- signal. Such a signal will be denoted by x(t)

• A signal that is specified only at discrete values of time is a

discrete-time signal. Such a signal will be denoted by x[nT] ,
x[n]

BITS Pilani, K K Birla Goa Campus

 Sinusoid in CT & DT
• Sinusoidal signal–
x(t)= A cos ω0t,t T0= 2π/ω0
• In the continuous-time case, the period
of a sinusoid can take on any value;
integral, fractional, or even irrational

• Sinusoidal sequence–
sequence
x[n] = cos ω0nT = cos Ωn
where Ω = ω0T
• The discrete-time signal, in contrast, is
specified only at integral values of n.
• so it is periodic only if Ω = 2π × some
rational number
BITS Pilani, K K Birla Goa Campus
 ANALOG, DIGITAL (qualifies
nature of y axis)
• A signal whose amplitude can take on any value in a
continuous range is an analog signal
signal.

• A digital signal, on the other hand, is one whose amplitude

can take on only a finite number of values.

BITS Pilani, K K Birla Goa Campus

Analogue vs Digital

Digital, continuous

Analogue, continuous

Digital, discrete

Analogue, discrete

BITS Pilani, K K Birla Goa Campus

 PERIODIC APERIODIC
PERIODIC,
• A signal x(t) is said to be periodic if and only if for some
positive constant T0 (T0 > 0)

• A di
discrete-time
t ti signal
i l is
i periodic
i di if and
d only
l if th
there exists
i t an
integer N0 > 0 such that

N0 is the period of x[n] in sample spacings.

• The smallest T0 or N0 is the fundamental period of the
periodic signal.

BITS Pilani, K K Birla Goa Campus

Example:

• Shifting x(t) by 1 time unit results in the same signal.

• Note - A signal is aperiodic if it is not periodic.

BITS Pilani, K K Birla Goa Campus

Properties of PERIODIC Signals
• A periodic signal x(t) remains unchanged when time-shifted
by one period.
period A periodic signal must start at t = - ∞ and
continue forever. Everlasting signals.
• Apperiodic signal
g x(t)
( ) can be g
generated byypperiodic extension
of any segment of x(t) of duration T0 (the period).

BITS Pilani, K K Birla Goa Campus

 Discrete Time Periodic Signal
• A discrete-time signal is periodic if and only if there exists an
integer N0 > 0 such that

N0 is the period of x[n] in sample spacings .

BITS Pilani, K K Birla Goa Campus

Notes

• Note that a sequence obtained by uniform sampling of a

periodic continuous-time signal may not be periodic .

• Note also that the sum of two continuous-time

continuous time periodic
signals may not be periodic but that the sum of two
periodic sequences is always periodic.

BITS Pilani, K K Birla Goa Campus

 Examples
1. Show that the complex exponential signal x(t) = ejωot is periodic
and that its fundamental period is 2π/ωo.
2. Let x(t) be the complex exponential signal x(t) = ejωo with angular
frequency ωo and fundamental period T0 = 2π/ωo.. Consider the
discrete-time sequence x[n] obtained by uniform sampling of x(t)
with sampling interval Ts. That is, x[n] = x(nTs) = ejωonTs. Find the
condition on the value of T. so that x[n] is periodic.
3 Let x1(t) and x2(t) be periodic signals with fundamental periods T1
3.
and T2, respectively. Under what conditions is the sum x(t) = x1(t)
+ x2(t) periodic, and what is the fundamental period of x(t) if it is
p
periodic?
4. Let x1[n] and x2[n] be periodic sequences with fundamental
periods N1 and N2, respectively. Under what conditions is the sum
x[n]
[ ] = x1[[n]] + x2[[n]] periodic,
p , and what is the fundamental period
p of
x[n] if it is periodic?

BITS Pilani, K K Birla Goa Campus

 Deterministic and Random
Deterministic
• A signal
i l whose
h physical
h i l
description is known completely,
either in a mathematical form or a
graphical form, is a deterministic
signal. SINUSOID
Random
• A signal whose values cannot be
predicted p
p precisely
y but are known
only in terms of probabilistic
description, such as mean value,
mean squared value,
value and so on is
a random signal. NOISE.
BITS Pilani, K K Birla Goa Campus
 Energy and Power Signals
• The signal which has finite energy and zero average power is
called energy signal.
signal The non periodic signals like exponential
signals will have constant energy and so non periodic signals
are energy signals.
• i.e., For energy signal, 0 < 𝐸 < ∞ 𝑎𝑛𝑑 𝑃 = 0
• For Continuous time signals,

• For Discrete time signals

signals,

BITS Pilani, K K Birla Goa Campus

 Energy and Power Signals

• The signal which has finite average power and infinite energy
i called
is ll d power signal.
i l Th
The periodic
i di signals
i l lik
like sinusoidal
i id l
complex exponential signals will have constant power and so
periodic signals
p g are p
power signals.
g
• i.e., For power signal, 0<𝑃<∞ 𝑎𝑛𝑑 𝐸=∞
• For Continuous time signals,

• For
F Discrete
Di t titime signals,
i l

BITS Pilani, K K Birla Goa Campus

 Exception
• Signals that satisfy neither property are referred to as neither
energy signals nor power signals
• Example:
• A Ramp signal x(t) = t ; t > 0 increases indefinitely as It t → ∞,,
and neither the energy nor the power exists for this signal.

• Therefore Ramp Signal is neither Energy nor Power Signal

Signal.

BITS Pilani, K K Birla Goa Campus

Practical Signals

• All practical signals have finite energies and are therefore

energy signals.
• A power signal must necessarily have infinite duration;
otherwise its power,
power which is its energy averaged over an
infinitely large interval, will not approach a (nonzero) limit.
• Clearly, it is impossible to generate a true power signal in
practice because such a signal has infinite duration and
infinite energy
• Also because of periodic repetition,
repetition periodic signals for
which the area under |x(t)|2 over one period is finite are
power signals; however, not all power signals are
periodic.
i di

BITS Pilani, K K Birla Goa Campus

 Causal and Non-causal signal
Continuous domain:
• Causal signal: A signal is said to be
causal if it is defined for t ≥ 0.

• Non-causal signal: A signal is said

to be non-causal,
non causal, if it is defined for
t < 0 or for both t < 0 and 𝑡 ≥ 0

• When a non-causal signal is

defined only for t < 0, it is called as
anti-causal signal

BITS Pilani, K K Birla Goa Campus

Causal and Non-causal signal

• An everlasting signal is always noncausal but a noncausal

signal
i l iis nott necessarily
il everlasting.
l ti
• A true everlasting signal cannot be generated in practice

BITS Pilani, K K Birla Goa Campus

Causal and Non-causal signal
Discrete domain:
• Causal
C l signal:
i l A signal
i l iis said
id tto b
be causal,
l if it iis d
defined
fi d
for n ≥ 0.
i.e., x[ n ]  0, for n  0
• Non-causal signal: A signal is said to be non-causal, if it is
defined for n < 0 or for both n < 0 𝑎𝑛𝑑 𝑛 ≥ 0
i.e., x[ n ]  0, for n  0

• Wh
When a non-causall signal
i l iis d
defined
fi d only
l ffor n < 0
0, it iis
called as anti-causal signal

BITS Pilani, K K Birla Goa Campus

Even and Odd

• An even signal is symmetric about the origin

x(t) = x(- t)
• An odd signal is antisymmetric about the origin
x(t)
( ) = - x(-
( t))

BITS Pilani, K K Birla Goa Campus

Decomposition in Even & Odd

• Any signal can be decomposed into even and odd

components
xe(t) = 1/2 [x(t) + x(- t)]
xo(t) = 1/2[x(t) - x(-
x( t)] :
Check that
xe(t) = xe((- t);
xo(t) = - xo(- t);
and that
xe(t) + xo(t) = x(t):

BITS Pilani, K K Birla Goa Campus

 Example

xe(t) =1/2 [x(t) + x(−t)] xo(t) =1/2 [x(t) − x(−t)]

• Same type of decomposition applies for discrete-time
discrete time
signals.
BITS Pilani, K K Birla Goa Campus
Some Useful Signal
Operations
• Amplitude Scaling
• Time Scaling
• Time Shifting
• Ti
Time Reversal
R l
• Combined Operations

BITS Pilani, K K Birla Goa Campus

Amplitude Scaling

• The scaled signal ax(t) is x(t) multiplied by the constant a

• The scaled signal ax[n] is x[n] multiplied by the constant a

BITS Pilani, K K Birla Goa Campus

Time Scaling, Continuous Time

• A signal x(t) is scaled in time by multiplying the time

variable by a positive constant b, to produce x(bt). A
positive factor of b either expands (0 < b < 1) or
compresses
p ((b > 1)) the signal
g in time.

BITS Pilani, K K Birla Goa Campus

Time Scaling, Discrete Time
• The discrete-time sequence x[n] is compressed in time by
multiplying the index n by an integer kk, to produce the time
time-
scaled sequence x[nk].
• This extracts every kth sample of x[n].
• I t
Intermediate
di t samples
l are lost.
l t
• The sequence is shorter.

• Called downsampling, or decimation.

BITS Pilani, K K Birla Goa Campus
Time Scaling, Discrete Time
• The discrete-time sequence x[n] is expanded in time by
dividing the index n by an integer m
m, to produce the time
time-
scaled sequence x[n/m].
• This specifies every mth sample.
• Th iintermediate
The t di t samples
l mustt be
b synthesized
th i d ((sett tto zero, or iinterpolated).
t l t d)
• The sequence is longer.

• Called upsampling, or interpolation.

BITS Pilani, K K Birla Goa Campus

Illustration of time scaling (down sampling) a
discrete-time signal by a factor of 2 and 3.
3

BITS Pilani, K K Birla Goa Campus

Illustration of time scaling (upsampling)
a discrete-time signal by a factor of 2
2.

BITS Pilani, K K Birla Goa Campus

Time Reversal
• Continuous time: replace t with -t, time reversed signal is x(-t)

• Discrete time: replace

p n with -n,, time reversed signal
g is x[-n].
[ ]

• Same as time scaling, but with b = -1.

BITS Pilani, K K Birla Goa Campus

Time reversal of a discrete-
time signal.
signal

BITS Pilani, K K Birla Goa Campus

Time Shifting Signals

Time shifting is an operation on a signal that shows up in

many areas of signals and systems:
– Time delays due to propagation of signals

• acoustic signals
g p
propagate
p g at the speed
p of sound
• radio signals propagate at the speed of light
– Time delays can be used to “build” complicated signals

• We’ll
W ’ll see thi
this llater
t
Time Shift: If you know x(t), what does x(t – t0) look like?
– For example…
p If t0 = 2:

x(0 – 2) = x(–2) x(1 – 2) = x(–1)

At t = 1, x(t – 2) takes the 
At t = 0, x(t – 2) takes the  value of x(t) at t = –1
value of x(t) at t = –2

BITS Pilani, K K Birla Goa Campus

Time Shift, Continuous Time

• For a continuous-time signal x(t), and a time t1 > 0,

• Replacing t with t - t1 gives a delayed signal x(t - t1)
• Replacing t with t + t1 gives an advanced signal x(t + t1)

• May seem counterintuitive.

BITS Pilani, K K Birla Goa Campus

Time Shift, Discrete Time

• For a discrete time signal x[n], and an integer n1 > 0

• x[n - n1] is a delayed signal.
• x[n + n1] is an advanced signal.
• Th delay
The d l or advance
d iis an iinteger number
b off sample
l times.
i

BITS Pilani, K K Birla Goa Campus

Time Shift, Discrete Time

BITS Pilani, K K Birla Goa Campus

Example of Time Shift of the
Unit Step u(t):

General View:
General View:
x(t ±
( t0)) for t
f 0 > 0
“+t0” gives Left 
shift (Advance)
hift (Ad )
“–t0” ggives Right 
g
shift (Delay)

BITS Pilani, K K Birla Goa Campus

 Combinations of Operations
• Time scaling, shifting, and reversal can all be combined.
• Operation
O ti can beb performed
f d in
i any order,
d b butt care iis
required.
• This will cause confusion.
• Example: x(2(t -1))
• Scale first, then shift
• Compress by 2 2, shift by 1

BITS Pilani, K K Birla Goa Campus

 Example x(2(t - 1)), continued
Shift first, then scale
Shift b
by 1
1, compress b
by 2 Incorrect

Shift first, then scale

Rewrite x(2(t - 1)) = x(2t - 2)
Shift byy 2,, scale byy 2 Correct

BITS Pilani, K K Birla Goa Campus

Elementary signals:

• Unit Impulse
• Unit Step
• Unit Ramp
• E
Exponential
i l

BITS Pilani, K K Birla Goa Campus

 The Impulse Function
Other Names: 
Delta Function,
One of the most important functions for Dirac Delta 
F ti
Function
understanding systems!!
Ironically…it does not exist in practice!!
It is
i a theoretical
th ti l ttooll used
d tto understand
d t d what
h t iis iimportant
t t
to know about systems!
But… it leads to ideas that are used all the time in practice!!
p
There are three views we’ll take of the delta function:
Rough View: a pulse with Infinite height, Zero 
width, Unit area
“A
A really narrow, really tall pulse that 
really narrow, really tall pulse that
has unit area”
BITS Pilani, K K Birla Goa Campus
Slightly Less-Rough View: Limit of
pulse with width ε and height 1/ε

So as ε
So as ε gets smaller
gets smaller the pulse gets higher
the pulse gets higher
and narrower but always has area of 1…

IIn the limit as ε gets smaller
th li it t ll it “becomes” 
it “b ”
the delta function

Precise Idea: δ(t) is defined by its behavior inside 
P i Id δ(t) i d fi d b it b h i i id
an integral:
The delta function δ(t) is defined as something that satisfies the 
The delta function δ(t) is defined as something that satisfies the
following two Conditions  (t )  0 for any t  0



  (t )dt  1 for any   0
BITS Pilani, K K Birla Goa Campus
Unit Impulse

• Obtaining a unit impulse as the limit of a pulse with unit

area. ((a)) The pulse q ((t),
) ((b)) a narrower and taller version
obtained by reducing a, (c) unit-impulse function obtained
as the limit.

BITS Pilani, K K Birla Goa Campus

Showing Delta Function on a
Plot:
We show δ(t) on a plot using an arrow…
(conveys infinite height and zero width)

BITS Pilani, K K Birla Goa Campus

Properties of impulse function

• There are many properties of δ(t)

1. sampling property:

2. sifting property:

BITS Pilani, K K Birla Goa Campus

The Sifting Property is the most
important property of δ(t):
t0  

 f (t ) (t  t )dt  f (t )  0
t0 
0 0

Integrating the product 
of f(t) & δ(t – to) returns 
a single number… the
a single number… the 
value of f(t) at the 
“location” of the 
shifted delta function
hift d d lt f ti

As long as the integral’s limits surround the “location” of the 
delta… otherwise it returns zero BITS Pilani, K K Birla Goa Campus
Illustration of the sampling property
of the unit-impulse function

BITS Pilani, K K Birla Goa Campus

Steps for applying sifting
property:

BITS Pilani, K K Birla Goa Campus

Time Scaling Property

BITS Pilani, K K Birla Goa Campus

Discrete-time unit impulse:

BITS Pilani, K K Birla Goa Campus

 Unit-step function
• The unit-step function is useful in situations where we need
to model a signal that is turned on or off at a specific time
instant. It is defined as follows

• A time shifted version of the unit-step function can be written

as

BITS Pilani, K K Birla Goa Campus

The use of the unit-step function for
switching a signal on at a specified time.
time

BITS Pilani, K K Birla Goa Campus

Relationship unit-step function
and the unit-impulse function
• The relationship between the unit-step function and the unit-
impulse function is important.
• The unit-step function can be expressed as a running
integral of the unit-impulse function:

• The limits of this integral vary with time but the integration is
always over a time interval of fixed length T .
• The unit-impulse function can be written as the derivative of
the unit-step function, that is,

BITS Pilani, K K Birla Goa Campus

Discrete-time unit step:

It can be shown that

BITS Pilani, K K Birla Goa Campus

Unit-pulse function
• We will define the unit-pulse function as a rectangular pulse
with unit width and unit amplitude
amplitude, centered around the
origin. Mathematically, it can be expressed as

BITS Pilani, K K Birla Goa Campus

Constructing a unit-pulse signal
using unit-step functions
functions.
• The unit-pulse function can be expressed in terms of time
shifted unit-step functions as

BITS Pilani, K K Birla Goa Campus

Unit Ramp Function r(t)

BITS Pilani, K K Birla Goa Campus

Constructing a unit-ramp
function

BITS Pilani, K K Birla Goa Campus

 Sinusoidal signals
• The general form of a sinusoidal signal is

• A is the amplitude of the signal

signal, ω0 is the radian frequency
which has the unit of radians per second, abbreviated as
rad/s. The parameter θ is the initial phase angle in radians.
Th radian
The di ffrequency ω0 can beb expressed d as ω0 = 2πf
2 f0
where f0 is the frequency in Hz.

BITS Pilani, K K Birla Goa Campus

 Periodicity of discrete-time
sinusoidal signals
• The general form of a discrete-time sinusoidal signal x[n] was given by:

which reveals the meaning of the normalized frequency F0 = f0 /fs

• For x[n] to be periodic,
periodic it needs to satisfy the periodicity condition
condition.
• Specifically we need
A cos (2πF0n + θ) =A cos (2πF0 [n + N] + θ)
=A cos (2πF0n + 2πF0N + θ)
• For above Eqn to hold, the arguments of the cosine functions must differ
by an integer multiple of 2π. This requirement results in
2πF0N = 2πk
and consequently N =k/F0 for the period N.
• Since we are dealing with a discrete-time signal, there is the added
q
requirement that the p
period N must be an integer
g value. Thus,, the discrete-
time sinusoidal signal is periodic provided it yields an integer value for N.

BITS Pilani, K K Birla Goa Campus

Continuous Exponential
Signals
• The continuous exponential signal is given by
x(t) = ceat
• Case 1: c and a real
• If a < 0 this is exponential decay
decay.
• If a > 0 this is exponential growth.

BITS Pilani, K K Birla Goa Campus

Continuous Exponential
Signals
• Case 2: c Real and a is purely imaginary
Let c=1 and a = 0 + jωo
x(t) = ejωot
x(t)
( ) is
i periodic?
i di ?
Check :
For periodicity it is required that
x(t + T) = e jωo(t +T) = e jωot e jωoT = x(t) = e jωot
This is true if e jωoT = 1 that is ωoT = 2πm (m is integer)
Fundamental Period T = 2π/ωo
Fundamental Frequency f = 1/T = ωo/2π

BITS Pilani, K K Birla Goa Campus

Continuous Exponential
Signals
• Case 3: C Real and a is Complex
Let c=A and a = = σ + jωo
x(t) = A e (σ + jωo)t = A eσt ejωot
Real Part of this Signal
Re{x(t)} = Re{ A eσt ejωot} = Re{A eσt (cosωot +j sinωot)}
Re{x(t)} = A eσt cosωot

BITS Pilani, K K Birla Goa Campus

Complex Plane

• Each complex frequency s = σ + jω corresponds to a

position in the complex plane.

BITS Pilani, K K Birla Goa Campus

Discrete Time Exponential (ejΩk)

BITS Pilani, K K Birla Goa Campus

Discrete Time Exponential (ejΩk)

• The magnitude and angle of ejΩk are 1 and kΩ, respectively.

• Therefore,
Therefore the points ej0, ejΩ, ej2Ω , ej3Ω , ... , ejkΩ, lie on a circle
of unit radius (unit circle) at angles 0, Ω , 2Ω, 3Ω, …,kΩ,
respectively.
• For
F each h unitit iincrease iin kk, th
the ffunction
ti ejΩk moves along l th
the
unit circle counterclockwise by an angle Ω.
• Therefore, the locus of ejΩk mayy be viewed as a p phasor
rotating counterclockwise at a uniform speed of Ω radians per
unit sample interval.
• The exponential e-jΩk , on the other hand hand, takes on values ej0,
e-jΩ, e-j2Ω , e-j3Ω , ... , e-jkΩ at k = 0, 1, 2, 3, . . . , k.
• Therefore, e-jΩk may be viewed as a phasor rotating clockwise
at a uniform speed of Ω radians per unit sample interval interval.

BITS Pilani, K K Birla Goa Campus

 Discrete Exponential signal
• Discrete Exponential signal is given by:

• Such a sequence can also be equivalently expressed in the

form

where β is a complex constant chosen as β = lnα. (This form is

more similar
i il tto th
thatt presented
t d ffor CT complex
l exponentials).
ti l )
• A complex exponential can exhibit one of a number of distinct
modes of behavior,
behavior depending on the values of the parameters
C and α.
• For example, as special cases, complex exponentials include
real exponentials and complex sinusoids.

BITS Pilani, K K Birla Goa Campus

Real Exponential (C and α are
real) α Positive

BITS Pilani, K K Birla Goa Campus

Real Exponential α Negative

(increasing cone) (decreasing cone)

BITS Pilani, K K Birla Goa Campus

 Complex Exponential
Sequence
• In the most general case of a complex exponential x(n) = cαn, c and
α are both complex.
• Letting c = |c|ejθ and α = |α|ejΩ where θ and Ω are real, and using
Euler’s relation, we can rewrite x(n) as
x(n)
( ) = |c||α|
| || |n cos(Ωn+
( θ)) + j ||c||α|
|| |n sin(Ωn+
( θ))
Re{x(n)} Im{x(n)}
• Thus, Re{x} and Im{x} are each the product of a real exponential
and real sinusoid.
sinusoid
• One of several distinct modes of behavior is exhibited by x,
depending on the value of α.
• If |α| = 1,
1 Re{x} and Im{x} are real sinusoids
sinusoids.
• If |α| > 1, Re{x} and Im{x} are each the product of a real sinusoid
and a growing real exponential.
• If |α|
| | < 1,
1 Re{x}
R { } and d Im{x}
I { } are each
h th
the product
d t off a reall sinusoid
i id
and a decaying real exponential.
BITS Pilani, K K Birla Goa Campus
General Complex
Exponentials
||α|| > 1 ||α|| < 1

|α| =1

BITS Pilani, K K Birla Goa Campus

 DT versus CT Exponentials
• Although there are many similarities between CT and DT
signals there are also important differences
signals, differences.
• One of these is the different properties of complex exponential
signals x(t) = ejω0t and x[n] = ejΩ0n
• We
W discussed
di d x(t)
(t) = ejω0t
j 0t previously
i l and d we can id
identify
tif ttwo
important properties of it :
• it is periodic for any value of ω0 and its fundamental period is T0 = 2π/ω0
• the larger the magnitude of ω0, the higher the rate of oscillation (i.e. frequency)
in the signals.
• Both of the above properties are different for x[n] = ejΩ0n.
• x[n]
[ ] = ejΩ0n is
i periodic
i di onlyl if Ω0 can be
b written
itt iin th
the fform Ω0 = 2π
2 X m/N
/N ffor
some integers N > 0, and m.
• x[n] = ejΩ0n does not have a continually increasing rate of oscillation as we
g
increase the magnitude of Ω0.
• In particular, x1[n] = ejΩ0n is equal to x2[n] = ej(Ω0+k2π)n, k is integer.

BITS Pilani, K K Birla Goa Campus

Example

BITS Pilani, K K Birla Goa Campus

 Example cont..
• A discrete-time sinusoid cosΩk can be obtained by sampling
a continuous
continuous-time
time sinusoid cosΩt at unit time interval T = 1;
that is, cosΩt sampled at t = 0, 1, 2, 3,…
• This fact means cosΩt is the envelope p of cosΩk.
• Since the period of cosΩt is 2π/Ω there are 2π/Ω , number of
samples (elements) of cosΩk in one cycle of its envelope.
• This
Thi number b may or may nott b be an iinteger
t
• In Fig cos(πk/4), has exactly 8 samples in each cycle of its
envelope (2π/Ω = 8).8) Thus,
Thus cos(πk/4) repeats every cycle of
its envelope. Clearly, cos(πk/4) is periodic with period 8

BITS Pilani, K K Birla Goa Campus

Example cont..

• On the other hand, cos (4πk/17), has an average of

approx 8.5 samples (not an integral number) in one cycle
of its envelope.
• Therefore,
Therefore the second cycle of the envelope will not be
identical to the first cycle.
• But there are 17 samples (an integral number) in 2
cycles of its envelope.
• Hence, the pattern becomes repetitive every 2 cycles of
its envelope.
envelope
• Therefore, cos (4πk/17) is also repetitive but its period is
17 samples (two cycles of its envelope).

BITS Pilani, K K Birla Goa Campus

 Example cont..
• If Ω/2π is irrational, it is impossible to fit an integral number
(No) of samples in an integral number (m) of cycles of its
envelope, and the pattern can never become repetitive.
• For instance, the sinusoid cos(0.8k)
( ) has an average g of 2.5
samples (an irrational number) per envelope cycle, and the
pattern can never be made repetitive over any integral
number (m) of cycles of its envelope;
• So cos(0.8k) is not periodic.

BITS Pilani, K K Birla Goa Campus

Example 2

BITS Pilani, K K Birla Goa Campus

Compare ejω0t in CT and DT

BITS Pilani, K K Birla Goa Campus

 Periodicity Properties of Discrete-time
Complex Exponentials

• Consider the discrete - time complex exponential with

f
frequency Ω0 + 2π:
2

ej(Ω0+2π)n = ejΩ0nej2πn = ejΩ0n

• From this we conclude that the exponential at frequency Ω0+
2π is same as at frequency
q y Ω0
• Similarly at frequencies Ω0 ± 2π, Ω0 ± 4π, and so on.
• This is very different from the continuous - time case whereby
the signals are all distinct for all distinct values of ω0
• Because of this periodicity of 2π, we need only to consider
frequency interval of 2π in the case for discrete - time signals

BITS Pilani, K K Birla Goa Campus

Periodicity Properties of Discrete-time
Complex Exponentials

BITS Pilani, K K Birla Goa Campus

Complex Sinusoidal Time
Function
• In many cases it is desirable to write a real-valued sinusoid
in terms of “complex
complex-valued
valued sinusoids”
sinusoids . This is a math trick
that – believe it or not! –makes things easier to work with!!!

BITS Pilani, K K Birla Goa Campus

Exploring the Complex
Sinusoidal Terms

BITS Pilani, K K Birla Goa Campus

 BITS Pilani
K K Birla Goa Campus

Thank You!