Signals & Systems

Prof. Dr. Vijaya C.

Department of Electronics & Communication Engineering
SDMCET, Dharwad
 Unit 3: Frequency domain
Representation
● Fourier Representations of Signals
– Introduction
– Fourier Series: DTFS, CTFS
– Fourier Transform: DTFT, CTFT
 Fourier Representations of Signals
• Represent a signal as a weighted superposition of
complex sinusoids.

• The weight associated with a sinusoid of a given

frequency represents the contribution of that
sinusoid to the overall signal.

• If such a signal is applied to an LTI system, then

the system output is a weighted superposition of
the system response to each complex sinusoid.
 Fourier Representations of Signals
• Joseph Fourier (1768-1830) developed the theory
of representations signals and systems using
sinusoids
– Fourier analysis
• Fourier methods have been used in every branch
of engineering and science.
• There are four distinct Fourier representations,
each applicable to a different class of signals
– Based on the periodicity properties of the signal and
– whether the signal is discrete or continuous in time.
 Four Fourier Representations of Signals
• Periodic signals : Fourier Series (FS) representations.
• Aperiodic signals: Fourier transform (FT) representation.
• Frequency domain representation of time domain signals
 Orthogonality property
• Before proceeding further, study of Orthogonality
property is necessary.

• Similarly in continuous time domain...

 Notations
• Small letter for a function implies time domain signal.
Capital letter implies Fourier representation.
• If an array of numbers are there with an arrow mark
below certain sample, it implies that it is the sample at
n=0 (or k=0).
– The samples to its left are the samples for –ve index of n (or
k) and
– The samples to its right are the samples for +ve index of n
(or k)
• If an array of numbers are there without an arrow
mark below certain sample, it implies that the first
sample in the array is the sample at n=0 (or k=0).
 Periodic Signals: FS Representations
• The DTFS representation of a periodic signal x[n]
with fundamental period N and fundamental
frequency is given by

• Where are the DTFS

coefficients of the signal x[n].

• x[n] and X[k] are a DTFS pair and denote this

relationship as
 Periodic Signals: FS Representations
• Either x(n) or X(k) provides complete information
about the signal.

• DTFS is a frequency domain representation for

x[n]
– Because each coefficient is associated complex
sinusoid of a different frequency.
– The variable k determines the frequency of the
sinusoid associated with X[k]
– And X[k] is function of frequency.
 Periodic Signals: FS Representations
• The DTFS is the only Fourier representation that
can be numerically evaluated and manipulated in
a computer.
– Since both signal and DTFS require finite number of
numbers to represent
 DTFS examples
• Follow the analysis relation to find the
coefficients
• or
• Employ Synthesis relation (representation of x(t)
in terms of FS coefficients) and determine FS.
 Determine DTFS of the signals shown

• (a)

• (b)
 Answer…
(a) x(n)={1 2 0 0 0 2}
• Thus N=6
• Apply analysis relation and orthogonality property ...
• X(k)={5, 3, -1, -3, -1, 3} /6
 Answer…
• (b) x(n)={0 1 2 0 0 0 0 0 0 0 0 0 0 -2 -1 } & N=15.
• Frequency is 2pi/15

• X(k)={ 0, - j3.786, - j5.464, - j4.253, - j1.157, j1.732, j2.629, j1.21,

- j1.21, - j2.629 , -j1.732, j1.157, j4.253, j5.464, j3.786}/15

• Observe the conjugate symmetry among the coefficients.

 DTFS examples
• Determine DTFS of the signal

• Answer: 14 can be written as 2x7.

• Similarly 10=2x5
• Now identify the Fundamental frequency and N: N=19,
Fundamental frequency =2pi/19
• Determine X(k)
• Verify if it is non zero at k=7, k=-7, k=5, k=-5 and k=0.
• X(k)= 1/j at k=7, X(k)= -1/j k=-7,
• X(k)= 0.5 at k=5, k=-5 and X(0)=1.
 DTFS examples
• Contd…detailed explanation…
• Given

• DTFS is determined by

• And by applying orthogonality property,

 DTFS examples
• Determine DTFS of the signal

• Answer:
x(n)={…2,0,-1,1,1,0,0,0,1,1,-1,0, 2,0, -1,1,1,0,0..}
N=12. Fundamental Frequency=2pi/12.
Determine DTFS…
 DTFS examples
• Use the definition of the DTFS to determine the
time domain signals represented by the following
DTFS coefficients: (determine inverse DTFS
without using that relation directly)
• (a)

• (b)
 DTFS examples
• Answer.. (a)

• From the definition of DTFS,

And for n = -10 from 10, signal x(n) is :

x(-4)= (21)0.5= 10.5, x(4)=(21)0.5 = 10.5
and for all other n, x(n) =0
 DTFS examples
• (b)

• Magnitude X(k)={0 0 0 1 1 0 0 0 0 0 1 1 0 0}
– Or magnitude X(k)={1 at k=3, 4, 10, 11}
• N=14. Fundamental frequency = 2pi/14=pi/7
• Phase X(k): = {-0.5pi at k= 4, 10 and 0.5pi at k= 3, 11}
 DTFS examples
• (b)contd…OR
• Magnitude X(k)={0 0 1 1 0 0 0 0 0 1 1 0 0 0}

• Phase X(k) = {-0.5pi at k= -4, 4 and 0.5pi at k= -3, 3}

• Compute x(n)
 Fourier representation example
• Consider two ECG
waveforms
• One is normal ECG and
the other is with
Ventricular tachycardia

• They are drawn as

continuous functions,
due to the difficulty of
depicting all 2000
values in each case.

• Apply suitable Fourier

representation to
identify the decease.
 DTFS example
• Ventricular tachycardia
– Rapid heart rate of
approximately 150 beats per
minute.
– Ventricular complexes are
wide about 160 ms in
duration
– The period N = 421.

• Normal ECG
– Impulse type
– heart rate of approximately
72 beats per minute.
– complexes less than 110 ms
– The period N = 305

• Both signals
– nearly periodic
– only slight variations in the
amplitude
– and length of each period.
 DTFS example
• Magnitude
spectrum DTFS
of one period of
each ECG is
shown.

• First 60 DTFS
coefficients is
depicted in Figs.
(c) and (d).
– The higher
indexed
coefficients are
very small and
thus are not
shown.
• The DTFS coefficients of
the normal ECG
DTFS example
– Approximately
constant,
– Gradual decrease in
amplitude as the
frequency increases.
– A fairly small
magnitude, since there
is relatively little power
in the impulsive signal.

• The DTFS coefficients of

ventricular tachycardia
ECG
– A greater dynamic
range,
– Larger amplitude
– The low frequency
coefficients containing
a large proportion of
the total power.
– Has greater power than
the normal ECG
 Continuous-time periodic signals: FS
• Continuous-time periodic signals are represented by the
Fourier series (FS).
• A signal x(t) with fundamental period T and fundamental
frequency is represented in terms of its FS
coefficients as

With as FS coefficients

• The signal and its FS pair is indicated as

 Continuous-time periodic signals: FS
• Thus, signal x(t) can be expressed as combination of
infinite number of analog frequency components, with
each of the kth component being given weight of X(k).
• These frequency components are analog frequency
components. And hence they are in infinite in number.
• Analog frequency components are measured in radians
/sec. (frequency in Hz=1/T)

• In the case of DTFS, the frequencies are digital

frequencies , ranging from 0 to 2pi radians.
• Digital frequency components are measured in radians.
• Any number in this range is a digital frequency.
 Continuous-time periodic signals: FS
• Determine FS of
• Answer: Follow the analysis relation to find the FS
coefficients
• or
• Employ representation of x(t) in terms of FS
coefficients and determine FS.
 Continuous-time periodic signals: FS
• Contd…

(k =-infinity to infinity)

Refer
…
 Continuous-time periodic signals: FS
• Determine FS of the following signal.

• Answer: T=1 time unit. Fundamental frequency

=2pi/T =2pi
• As indicated in the waveform, even for –ve time
values, signal x(t) = . Thus it can be treated as
•
 Continuous-time periodic signals: FS
• Contd.. check if integration steps are correct…
 Continuous-time periodic signals: FS
• Determine inverse FS of the following FS

• Answer:
• Fundamental frequency is given = pi. Substituting X(k), in
x(t),

• Each delta function has a value of 1 only when its

argument is zero. For the first term, for example, k=2, for
the 2nd term, k=-2 and so on. Substituting,
 Continuous-time periodic signals: FS
• Determine inverse FS of the following FS, with
fundamental frequency =2pi
 Continuous-time periodic signals: FS
• Answer: Given fundamental frequency =2pi
• Magnitude X(k)={1 ; k=-4 to 4} &
• Phase angle X(k)={2pi, 1.5pi, pi, 0.5pi, 0, -0.5pi, -pi, -1.5pi, -2pi}

• Further simplify…
 Discrete Time Nonperiodic Signals:
The Discrete Time Fourier Transform
• The DTFT : Represents a
discrete-time nonperiodic
Synthesis or signal x(n) as a
Inverse DTFT: superposition of complex
IDTFT sinusoids.
– Nonperiodic signal x(n)
Analysis or ranges from –infinity to
Frequency domain infinity
representation – It can even have only
finite number of
of x[n]: samples.
DTFT – It is not periodic
– Thus variable of sum ‘n’
is –infinity to infinity
in analysis relation.
 Discrete Time Nonperiodic Signals:
The Discrete Time Fourier Transform
• Frequency component
is continuous variable.
Synthesis or – Thus, there is
Inverse DTFT: integration in synthesis
IDTFT relation
Analysis or • Frequencies are in the
Frequency domain interval -pi to pi or 0 to
representation 2pi radians
of x[n]: – Thus, the variable of
DTFT integration in
synthesis relation is
from -pi to pi
radians.
 Discrete Time Nonperiodic Signals:
The Discrete Time Fourier Transform
• The DTFT : Represents a discrete-time nonperiodic signal x(n)
as a superposition of complex sinusoids.
– Nonperiodic signal x(n) ranges from –infinity to infinity
– It can even have only finite number of samples.
– It is not periodic
– Thus variable of sum ‘n’ is in analysis relation

• Frequency component is continuous variable.

– Thus, there is integration in synthesis relation
• Frequencies are in the interval -pi to pi or 0 to 2pi radians
– Thus, the variable of integration in synthesis relation is from
radians.
 The Discrete Time Fourier Transform
• Use the defining equation for the DTFT to evaluate the
frequency-domain representations of the following
signals:

• Answer: (a)
 The Discrete Time Fourier Transform
• Use the defining equation for the DTFT to evaluate the
frequency-domain representations of the following
signals:

• Answer: (b) -ve sign makes the power of a always +ve

ge
an
ch
n
sig

Because n=0 is included

in summation.
 The Discrete Time Fourier Transform
• Use the defining equation for the DTFT to evaluate the
frequency-domain representations of the signal shown:

• Answer:
 The Discrete Time Fourier Transform
• Determine the time-domain signals corresponding to
the following DTFTs: (IDTFT...)
• (a)
• (b)

• Answer: (a)
 The Discrete Time Fourier Transform
• (a) contd…

• Apply orthogonality property to the first two terms

• When n=-1, the first term reduces to

• Since it occurs at n=-1 and for all other values of n it is

zero, it has to be written as
 The Discrete Time Fourier Transform
• (a) contd…
• Similarly, when n=1, the second term reduces to

• Since it occurs at n=1 and for all other values of n it is

zero, it has to be written as
 The Discrete Time Fourier Transform
• (a) contd…with third and fourth term,
 The Discrete Time Fourier Transform
• (a)
• (b)

1
• (b)
 The Discrete Time Fourier Transform
• Contd (b)
CT Nonperiodic Signals: Fourier Transform
• The continuous time domain signal is nonperiodic.
• Represented as superposition of infinite number
of complex sinusoids.
• This analog frequency range from
radians /sec
inverse FT , IFT or
synthesis relation

FT or analysis relation
CT Nonperiodic Signals: Fourier Transform
• The continuous time domain signal is nonperiodic.
• The Fourier Transform describes the signal x(t) as
a function of frequency and is termed the
frequency domain representation of x(t).
• The inverse FT maps the frequency domain
representation back into the time domain.
• Verify the following:
CT Nonperiodic Signals: Fourier Transform
• Determine FT of the following signals

• Answer:
CT Nonperiodic Signals: Fourier Transform

• Answer:

• Applying sifting property,

Thus, signal x(t)=delta

function has all the
1
frequency components
and to the same extent
=1.
The Four Fourier Representations
• Continuous or discrete
time periodic signals have
a series representation.
• In this, the signal is
represented as a weighted
sum of complex sinusoids
of frequencies which are
integer multiple of the
fundamental frequency of
the signal.
• A discrete set of
frequencies, X(k), is
involved in the series.
• Hence, the frequency
domain representation
involves a discrete set of
weights or coefficients.
 The Four Fourier Representations
• Nonperiodic signals, both
continuous and discrete
time have Fourier
Transform
representations.
• It involves weighted
integrals of complex
sinusoids over a
continuum of frequencies.
• Accordingly, the frequency
domain representation for
nonperiodic signals is a
continuous function of
frequency.
 The Four Fourier Representations
• Also observe that the Fourier representations of discrete time
signals, either the DTFS or the DTFT, are periodic functions of
frequency.
• This is because the discrete time complex sinusoids used to
represent discrete time signals are 2pi periodic functions of
frequency.
– That is, discrete-time sinusoids whose frequencies differ by integer
multiples of 2pi are identical.
 The Four Fourier Representations

• Fourier representations of continuous-time signals

involve superpositions of continuous-time sinusoids.
• Continuous-time sinusoids with distinct frequencies
are always distinct.
– Thus, the frequency domain representations of
continuous time signals are nonperiodic.
 The Four Fourier Representations
• Summarizing, discrete time signals have periodic frequency
domain representations.
• Continuous time signals have nonperiodic frequency domain
representations.
• In general, representations that are continuous in one domain are
nonperiodic in the other domain.
• Conversely, representations that are discrete in one domain are
periodic in the other domain.
The Four Fourier Representations
 FT pairs
FT

FT

FT
Unit III ends here