SESHADRI RAO GUDLAVALLERU ENGINEERING COLLEGE

(An Autonomous Institute with Permanent Affiliation to JNTUK, Kakinada)

Seshadri Rao Knowledge Village :: Gudlavalleru -521356

Academics Strengthening & Advancement (AS&A)

Department of Electronics and Communication

Engineering

R – 20
III B.Tech. I Semester

DIGITAL SIGNAL PROCESSING

Learning Material
2024-25

Prepared by: Faculty of ECE

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

BASIC ELECTRICAL AND ELECTRONICS ENGINEERING

(Common to All Branches)

III YEAR -I SEM

Lecture:3 Internal Marks :30

Credits:3 External Marks :70

Course Objectives

• To familiarize with the basic concepts of discrete time signals and systems

• To introduce the concepts of Z-transform and frequency domain

representation of discrete time signals.

• To familiarize with the designing of digital filters and their realization.

Course Outcomes

Upon successful completion of the course, the students will be able to

CO1.analyze and process signals in the discrete domain.

CO2.determine the Fourier series coefficients and z-transform of discrete time signals.

CO3.apply the various transform techniques on discrete time signals.

CO4.design digital filters (IIR and FIR) for a given specifications.

CO5.apply various windowing techniques in the design of FIR filter.

CO6.realize digital filters (IIR and FIR).

UNIT - I: Discrete Time Signals and System

Discrete time signals- classification, elementary discrete time signals, basic operations on
sequences; discrete time systems-classification, discrete time linear Time Invariant systems
and their properties, convolution sum.

UNIT - II: Z-Transform and Discrete Fourier Series

Z Transform of sequence, properties of ROC, properties of Z transform, inverse Z

transform- partial fraction method.

Discrete Fourier series: Fourier series for discrete time periodic signals, Fourier
Transform for discrete time non-periodic signals, energy density spectrum, relationship of
Fourier transform to Z transform, frequency response

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT - III: Discrete Fourier Transform

Frequency sampling- Discrete Fourier Transform (DFT), properties of DFT, linear

convolution of sequences using DFT, relationship between DFT and Z transform.

UNIT - IV: Fast Fourier Transforms (FFT)

Fast Fourier Transform-Radix-2 decimation in time and in frequency FFT algorithms,

IDFT using FFT algorithms.

UNIT - V: Design of IIR Filters

Analog filter approximation-Butterworth and Chebyshev (Type-I) filters, design of IIR

filters from analog filters- Impulse Invariant technique, Bilinear transformation

UNIT - VI: Design of FIR Filters

Linear Phase FIR filters-frequency response, Fourier Series method of designing FIR filter,
design of FIR filters using windows (Rectangular, Bartlett, Hamming, Hanning)

Realization of Digital Filters: Realization of IIR Filters- Direct form I, II; realization of
FIR filters- transversal structure, cascade realization

Text Books:

John G. Proakis, Dimitris G. Manolakis, “Digital Signal Processing, Principles, Algorithms, and
Applications”, Pearson Education / PHI, 2013.

Reference Text Books:

1. A.V.Oppenheim and R.W. Schaffer, “Discrete Time Signal Processing”, PHI.

2. Andreas Antoniou, “Digital Signal Processing”, TATA McGraw Hill, 2006.

3. MH Hayes, “Digital Signal Processing”, Schaum’s Outline series, TATA Mc-Graw Hill, 2007.

URLs and Other E-Learning Resources:

www.ieeexplore.ieee.org/servlet/opac?punumber97

www.ieeexplore.ieee.org/servlet/opac?punumber79

Digital Learning Materials:

Digital signal processing by Prof.S.C. Dutta Roy, IIT Delhi

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT I
Discrete Time Signals and Systems
Syllabus
Introduction to Digital Signal Processing, Discrete time signals- Classification, Elementary
discrete time signals, Basic operations on Sequences; Discrete time Systems-Classification,
Discrete time Linear Time Invariant Systems and their Properties.
Introduction to Digital Signal Processing
What is DSP?
DSP, or Digital Signal Processing, as the term suggests, is the processing of signals by
digital means. A signal in this context can mean a number of different things. Historically the
origins of signal processing are in electrical engineering, and a signal here means an electrical
signal carried by a wire or telephone line, or perhaps by a radio wave. More generally, however,
a signal is a stream of information representing anything from stock prices to data from a
remote-sensing satellite.
Analog and digital signals
In many cases, the signal is initially in the form of an analog electrical voltage or
current, produced for example by a microphone or some other type of transducer. In some
situations the data is already in digital form - such as the output from the readout system of a
CD (compact disc) player. An analog signal must be converted into digital (i.e. numerical) form
before DSP techniques can be applied. An analog electrical voltage signal, for example, can be
digitized using an integrated electronic circuit (IC) device called an analog-to-digital converter
or ADC. This generates a digital output in the form of a binary number whose value represents
the electrical voltage input to the device.
Signal processing
Signals commonly need to be processed in a variety of ways. For example, the output
signal from a transducer may well be contaminated with unwanted electrical "noise". The
electrodes attached to a patient's chest when an ECG is taken measure tiny electrical voltage
changes due to the activity of the heart and other muscles. The signal is often strongly affected
by "mains pickup" due to electrical interference from the mains supply. Processing the signal
using a filter circuit can remove or at least reduce the unwanted part of the signal. Increasingly
nowadays the filtering of signals to improve signal quality or to extract important information
is done by DSP techniques rather than by analog electronics.
Development of DSP

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The development of digital signal processing dates from the 1960's with the use of
mainframe digital computers for number-crunching applications such as the Fast Fourier
Transform (FFT), which allows the frequency spectrum of a signal to be computed rapidly.
These techniques were not widely used at that time, because suitable computing equipment
was available only in universities and other scientific research institutions.
Digital Signal Processors (DSPs)
The introduction of the microprocessor in the late 1970's and early 1980's made it
possible for DSP techniques to be used in a much wider range of applications. However,
general-purpose microprocessors such as the Intel x86 family are not ideally suited to the
numerically-intensive requirements of DSP, and during the 1980's the increasing importance
of DSP led several major electronics manufacturers (such as Texas Instruments, Analog
Devices and Motorola) to develop Digital Signal Processor chips - specialized microprocessors
with architectures designed specifically for the types of operations required in digital signal
processing. (Note that the acronym DSP can variously mean Digital Signal Processing, the
term used for a wide range of techniques for processing signals digitally, or Digital Signal
Processor, a specialized type of microprocessor chip). Like a general-purpose microprocessor,
a DSP is a programmable device, with its own native instruction code. DSP chips are capable
of carrying out millions of floating point operations per second, and like their better-known
general-purpose cousins, faster and more powerful versions are continually being introduced.
Applications of DSP
DSP technology is nowadays commonplace in such devices as mobile phones,
multimedia computers, video recorders, CD players, hard disc drive controllers and modems,
and will soon replace analog circuitry in TV sets and telephones. An important application of
DSP is in signal compression and decompression. In CD systems, for example, the music
recorded on the CD is in a compressed form (to increase storage capacity) and must be
decompressed for the recorded signal to be reproduced. Signal compression is used in digital
cellular phones to allow a greater number of calls to be handled simultaneously within each
local "cell". DSP signal compression technology allows people not only to talk to one another
by telephone but also to see one another on the screens of their PCs, using small video cameras
mounted on the computer monitors, with only a conventional telephone line linking them
together.
Although the mathematical theory underlying DSP techniques such as Fast Fourier and
Hilbert Transforms, digital filter design and signal compression can be fairly complex, the
numerical operations required to implement these techniques are in fact very simple, consisting

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

mainly of operations that could be done on a cheap four-function calculator. The architecture
of a DSP chip is designed to carry out such operations incredibly fast, processing up to tens of
millions of samples per second, to provide real-time performance: that is, the ability to process
a signal "live" as it is sampled and then output the processed signal, for example to a
loudspeaker or video display. All of the practical examples of DSP applications mentioned
earlier, such as hard disc drives and mobile phones, demand real-time operation.
The major electronics manufacturers have invested heavily in DSP technology.
Because they now find application in mass-market products, DSP chips account for a
substantial proportion of the world market for electronic devices. Sales amount to billions of
dollars annually, and seem likely to continue to increase rapidly.
Advantages and disadvantages of DSP
Advantages to using DSP techniques are:
• Reproducibility,
• Programmability (flexibility). DSP is flexible since the digital processing can often be easily
modified by programming,
• Stability and high reliability: Absence of component drift problems allows for complex
processing than is possible with analogue circuitry. DSP provides better signal quality and
repeatable performance resulting in lower costs for equivalent performance.
DSP techniques are limited, at present, to signals with relatively low bandwidths (5
MHz video bandwidth). The point at which DSP becomes too expensive will depend on the
application and the current state of the processing technology. The cost of high-speed ADC and
DAC devices and the extra circuitry required to implement high-speed designs, makes DSP
impractical and uneconomical for many applications such as simple filters. Higher power
consumption and size of a DSP implementation may make it unsuitable for small- size
applications.

Basic block diagram for a DSP system

Figure 1: DSP Block Diagram

• An anti-aliasing low-pass filter.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

• An A/D converter whose sampling period T = 1/fs. Here fs is the sampling frequency with f s
≥ 2fm. Here fm is the highest frequency present in the input signal. Aliasing will occur if the
sampling frequency is less than twice the highest frequency contained in the signal.
• This is the digital signal processor
• D/A Converter
• Reconstruction filter
1.1 Discrete Time Signals Classification and Elementary Discrete Time Signals

Basic Elementary Signals

1. Unit Step
The unit step signal is normally given the symbol u(n)
u(n) = 1 for n ≥0
u(n) = 0 for n<0
Note the signal is not zero before time = 0 but is just not defined at that time period.

2. Unit Impulse
It is defined as
δ(n) = 1 for n = 0
δ(n) = 0 elsewhere

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

It is important to understand that the impulse only exists at n = 0. The unit impulse δ(n),
may be expressed in terms of the step function as:
δ(n) = u(n) - u(n - 1)
3. Decaying exponential
G(n) = an for n ≥0
G(n) = 0 for n< 0
Where 0 < a < 1 i.e. a is a fractional number

4. Discrete Sinusoidal Signal

The analogue sine signal is
x(n) = A sinωan = A sin2πfan

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Classification of signals
1. Periodic and Aperiodic signals
Periodic signal:
• Given x(n) is a discrete-time signal
• x (n) is periodic if x(n) = x(n+N) for any N.
• Example
– x(n) = A cos(wn)
– x(n+N) = A cos[w(n+N)] = A cos(wn+wN)= A cos(wn+2π) = A cos(wn)
Aperiodic/ Non-periodic signal:
• For non-periodic signals
x(n) ≠ x(n+N)
• A non-periodic signal is assumed to have a period T = ∞
• Example of non-periodic signal is an exponential signal.
2. Energy and Power Signals:
Energy Signal:
• A signal with finite energy and zero power is called Energy Signal i.e.for energy signal
0 < E < ∞ and P =0
• Energy of a signal is defined as the area under the square of the magnitude of the signal.

Power Signal:

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

• Some signals have infinite signal energy. In that case it is more convenient to deal with
average signal power.
• For power signals
0<P<∞ and E = ∞

3. Deterministic & Non Deterministic Signals

Deterministic signals:
• Behavior of these signals is predictable w.r.t time
• There is no uncertainty with respect to its value at any time.
• These signals can be expressed mathematically.
For example x(n) = sin(3n) is deterministic signal.

Non Deterministic or Random signals

• Behavior of these signals is random i.e. not predictable w.r.t time.
• There is an uncertainty with respect to its value at any time.
• These signals can’t be expressed mathematically.
• For example Thermal Noise generated is non deterministic signal.

4. Even and Odd Signals

Even Signal

A signal is said to be even or symmetric if it satisfies the following condition;

x(−n)=x(n)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Here, we can see that x(-1) = x(1), x(-2) = x(2) and x(-n) = x(n). Thus, it is an even signal.

Odd Signal

A signal is said to be odd if it satisfies the following condition;

x(−n)=−x(n)

From the figure, we can see that x(1) = -x(-1), x(2) = -x(2) and x(n) = -x(-n). Hence, it
is an odd as well as anti-symmetric signal.

Assignment cum Tutorial Questions

Descriptive:
1. BL-II
2.
1.2 Basic Operations on Sequences
1. Amplitude scaling:

It is defined by the equation

y[n] = c*x[n]

Where x(n) is the input signal and c is the amplification factor.

Eg: An amplifier is a physical device which performs amplification of the signal.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2. Addition:

It is defined by the equation

y[n] = x1[n] + x2[n]

The values of x1[n] and x2[n] are added at every instant n.

Eg: An audio mixer is an example of an adder. It adds the voice and music signals.

3. Multiplication:

It is defined by the equation

y[n] = x1[n] * x2[n]

Eg: A DSBSC modulation scheme is an example for multiplication of signals. In

DSBSC, the carrier signal is multiplied with the message signal before transmission.

4. Time scaling:

It is defined as

y[n] = x[an]

Where ‘a’ is an integer

5. Reflection:

It is defined by the equation

y[n] = x[-n]

It is a signal which is a reflected version of the input signal about the amplitude axis.

6. Time shifting:

It is defined by the equation

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

y[n] = x[n – n0]

Where n0 is an integer.
Assignment cum Tutorial Questions
Objective:
1. Let x(n) be a signal which is zero for n < -2 and n > 8. Which of the following is
guaranteed to be zero for n < -3 and n > 12 BL-
III
a) x(n+4) b) x(-n-4) c) x(n-4) d) x(-n+4)
2. What is the relation between x(n) and y(n) depicted in figure? BL-
III

a) y(n)=x(2n-1) b) y(n)=x(n/2-1) c) y(n)=x(2n+1) d) y(n)=x(n/2+1)

3. What is the relation between x(n) and y(n) depicted in figure? BL-
III

a) y(n)=x(3n-3) b) y(n)=x(n/3-3) c) y(n)=x(3n+3) d) y(n)=x(n/3+3)

4. Time scaling operation with 2 units is also known as ___________
BL-II
a) Down-sampling b) Up-sampling c) Sampling d) Interpolation
5. Which property does y(n)=x(1-n) exhibit?
BL-I
a) Time scaling b) Time shifting
c) Reversal d) Time shifting and reversal
6. If x(n)={0,0,1,1,1,1,1,0} then x(3n+1) is? BL-
III
a) {0,1,0,0,0,0,0,0} b) {0,0,1,1,1,1,0,0}
c) {1,1,0,0,0,0,0,0} d) None of the mentioned
Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

7. Given a sequence x(n), to generate the sequence y(n) = x(3 − 4n), which one of the
following procedures would be correct ? Gate-08 BL-
III
a) First delay x(n) by 3 samples to generate z1(n), then pick every 4th sample of
z1(n) to generate z2(n), and then finally time reverse z2(n) to obtain y(n).
b) First advance x(n) by 3 samples to generate z1(n), then pick every 4th sample of
z1(n) to generate z2(n), and then finally time reverse z2(n) to obtain y(n)
c) First pick every fourth sample of x(n) to generate v1(n), time-reverse v1(n) to
obtain v2(n), and finally advance v2(n) by 3 samples to obtain y(n)
d) First pick every fourth sample of x(n) to generate v1(n), time-reverse v1(n) to
obtain v2(n), and finally delay v2(n) by 3 samples to obtain y(n)
Descriptive:
1. What are the different basic operations we can perform on the signals? Explain

BL-II
2. Perform x(-2n-1) on the signal x(n) = u(n+2) * u(-n+3).
BL-III
3. A discrete time signal is given as x(n)={1, 2, 3, 4, 5, 4, 3, 2, 1}. Represent the signals
x(2n) and x(n/2) in graphical representation. BL-
III
4. If x(n)={0,0,1,1,1,1,1,0} then find x(n+3). BL-
III
1.3 Discrete Time Systems Classification
➢ Static (Memory Less) and Dynamic (Memory)
➢ Linear and Non-Linear
➢ Time-Variant (Shift-variant) and Time-invariant (Shift-invariant)
➢ Causal and Non-causal
➢ Stable and Unstable
1. Static (Memory Less) and Dynamic (Memory) Systems
A system is said to be Static output depends only upon the present values of the input
but not past and future value of the input at any instant of time.
Examples: y (n) = sin{x(n)}
A system is said to be Dynamic output depends upon the past and future values of the
input but not the present value of the input at any instant of time.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Examples: y (n) = sin{x(2n)

2. Linear and Non-Linear Systems
A System is said to be Linear it must be satisfies the superposition (Additivity +
Homogeneity) principal otherwise the system is said to be Non-Linear system
A system is said to be linear if it is satisfies the following condition
T{a x1 (n) + b x2 (n) } = aT{x1 (n)} + b T{x2 (n)}
Examples: y(n) = sin{x(n)} [Non-Linear System]
y(n) = x(t)sinn [Linear System]
3. Time-Variant (Shift-variant) and Time-invariant (Shift-invariant)
If the input/output characteristic of systems does not vary with time then the system is
said to be Time-invariant System otherwise the system is said to be Time-Variant System.
Let x(n) be the input signal and y(n) be the output signal then the condition for Time-invariant
is
𝑇{𝑥(𝑛 − 𝑁)} = 𝑦(𝑛 − 𝑁)
Examples: y(n) = 2x(n) [Time-invariant]
y(n) = x(2n) [Time-Variant]
4. Causal and Non-causal
A system is said to be Causal when the output depends on the present and past values
of the input but not on the future values of the input otherwise the system is said to be Non-
Causal.
Examples: y(n) = x(n) + x(n-1) [Causal]
y(n) = x(n+2) [Non-Causal]
5. Stable and Unstable
A system is said to be stable if and only if every bounded input produces a bounded
output (BIBO). Otherwise the system is said to be unstable.
Examples: y(n) = r(n) x(n) [Unstable]
y(n) = e^-2n x(n) [Stable]
Where x(n) =u(n)
Assignment cum Tutorial Questions
Objective:
1. A causal system is physically
BL-I
a) Realize b) Reliable c) Invertible d) None

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2. The system described by the input-output equation y(n)=nx(n)+bx3(n) is BL-

II
a) Static system b) Dynamic system
c) Identical system d) None of the mentioned
3. If the output of the system of the system at any ‘n’ depends only the present or the
past values of the inputs then the system is said to be __________
BL-I
a) Linear b) Non-Linear c) Causal d) Non-causal
4. If a system do not have a bounded output for bounded input, then the system is said
to be __________
BL-I
a) Causal b) Non-causal c) Stable d) Non-stable
5. If a signal x(n) is processed through a system to obtain the signal x2(n), then the
system is said to be ____________
BL-II
a) Linear b) Non-linear c) Exponential d) None of the mentioned
6. What is the physical device that performs an operation on the signal?
BL-I
a) Signal source b) System c) Medium d) None of the mentioned
7. Consider a single input single output discrete-time system with x(n) as input and
y(n) as output, where the two are related as
𝑛|𝑥(𝑛)| 𝑓𝑜𝑟 0 ≤ 𝑛 ≤ 10
𝑦(𝑛) = {
𝑥(𝑛) − 𝑥(𝑛 − 1) 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Which one of the following statement is true about the system? Gate-17
BL-V
a) It is casual and stable b) It is casual but not stable
c) It is not casual but stable d) It is neither casual nor stable
8. A system is defined by its impulse response h(n) = 2n u(n − 2). The system is
Gate-11 BL-
III
a) stable and causal b) causal but not stable
c) stable but not causal d) unstable and non-causal
Descriptive:
1. How can you classify discrete time Systems? Explain. BL-
IV

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2. Check whether the following systems are static, Time In-variant, Causal and Linear.

BL-V
(i) y(n)= x(n) x(n-2) (ii) log10|x(n)| (iii) y(n)= an u(n)
3. State whether following systems are linear, causal, and time-invariant and stability.

BL-V
(i) y(n) = 2 x(n+1) + [x(n-1)]2 (ii) y(n) + y(n-1) = x(n) + x(n-2)
4. Check whether the following systems are causal and Stable.
BL-V
(i) h(n) =log |x(n)| (ii) h(n)= Cos nπ/2
1.4 Convolution and LTI Systems

A discrete time system is an algorithm that maps a discrete signal or input signal x(n)
onto a discrete output signal y(n). For a system to be linear:
a1 x1(n) + a2 x2(n) H →b1y1(n) + b2 y2(n)
(No extra terms generated by the system and is therefore linear) where a1, a2,, b1, b2 =
arbitrary constants. A time-invariant system is defined as
𝑇{𝑥(𝑛 − 𝑁)} = 𝑦(𝑛 − 𝑁)
Response of LTI System

For discrete time the representation takes the form of the convolution sum, while it’s
continuous time counterpart is the convolution integral.
For Discrete time:
+∞

y[n]= x[n]*h[n] = ∑ x[k] h[n-k] (1)

k=-∞

From above expressions we can conclude that the characteristics of an LTI system are
completely determined by its impulse response.
1. The Commutative Property:
A basic property of the convolution in both continuous and discrete time is that it is a
commutative operation.

x[n]*h[n]=h[n]*x[n]= ∑ h[n] x[n-k]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

From the above expressions, the output of an LTI system with input x[n] and unit
impulse response h[n] is identical to the output of an LTI system with input h[n] and unit
impulse response x[n].

2. The Distributive Property:

x[n]*(h1[n]+h2[n])= x[n]*h1[n]+x[n]*h2[n]

The distributive property has a useful interpretation in terms of system

interconnections. Consider two continuous time LTI systems in parallel a shown in figure. The
systems shown in the block diagram are LTI system with the indicated unit impulse responses.

h1(n)
h1(n)+h2(n) y(n)
y(n) x(n)
x(n)
h2(n)

(a) (b)

The two systems with impulse responses h1(n) and h2(n), have identical inputs and their
outputs are added. So,
y1(n) =x(n)*h1(n)

and y2(n) =x(n)*h2(n),

The system of fig (a) has output,

y(n) =x(n)*h1(n)+x(n)*h2(n),

And the system of figure (b) has output,

y(n) =x(n)*[h1(n)+h2(n)]

Hence, by the virtue of the distributive property of convolution, a parallel combination

of LTI systems can be replaced by single LTI system whose unit impulse response is the sum
of the individual unit impulse responses in the parallel combination.

3. The Associative Property:

It is an important and useful property of convolution.

x[n]*(h1[n]*h2[n])=(x[n]*h1[n])*h2[n]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

From the block diagram shown, an interpretation of the associative property can be
illustrated. In the Figure (a)

y[n] = w[n]*h2[n]
= (x[n]*h1[n])*h2[n]
In the figure (b)
y[n] = x[n]*h[n]
= x[n]*(h1[n]*h2[n])

x[n] h1[n] h2[n] y[n]

w[n]

(a)

h[n]=h1[n]+h2[n]
x[n] y[n]

(b)

According to the associative property, the series interconnection of the two systems in
fig(a) is equivalent to the single system in fig(b).This can be generalized to an arbitrary number
of LTI systems in cascade. This interpretation and conclusion also hold in continuous time.
From Figures we can conclude that the impulse response of the cascade of two LTI systems is
the convolution of their individual impulse responses.

The unit impulse response of a cascade of two LTI systems does not depend on the
order in which they are cascaded. The order in which they are cascaded does not matter as far
as the overall system impulse response is concerned.

4. LTI systems with or without memory:

A system is memoryless if its output at any time depends only on the value of the input
at the same time.

For discrete time LTI system, if h[n] =0 for n#0.in this case the impulse response has
the form

h[n]=Kδ[n],
Where K=h[0] is a constant and the convolution sum reduces to the relation y[n]=Kx[n]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

If a discrete time LTI system has an impulse response h[n] that is not identically zero
for n#0, then system has memory.

5. Causality for LTI systems

A system is said to be causal if the output at any time depends on values of the input at the
present time and in the past. If two inputs to a casual system are identical up to some point in
time to or no, the corresponding outputs must also be equal up to this same time.
In order for the discrete time to LTI systems to be casual, y[n] must not depend on x[k]
for k>n. We have,
From above equation, for this to be true, all the coefficients h[n-k] that multiply values of x[k]
for k>n must be zero. Then the impulse response of a casual discrete time LTI system satisfy
the condition

h[n] =0 for n<0

Hence, the impulse response of a casual LTI system must be zero before the impulse occurs.

6. Stability of the LTI systems:

A system is stable if every bounded input produces a bounded output. In order to determine the
conditions under which LTI systems are stable.

│x[n]│<B for all n

+∞

│y[n]│=│∑ h[k] x[n-k]│

k=-∞

+∞

≤ ∑│h[k] ││x[n-k]│
K=-∞

+∞

≤ B ∑│h[k] │
K=-∞

Therefore the system is stable if the impulse response is absolutely summable, i.e. if
+∞

∑ │h[k] │ < ∞
K=-∞

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Invertibilty of LTI system:

A system is invertible only if an inverse system exists when connected in series with
the original system and produces an output equal to the input to the first system. Consequently,
if LTI system is invertible then it has an LTI inverse.

Let us consider a discrete time system shown in the figure below with impulse response
h1(n) which results in x(n) such that series interconnection in figure(a) is identical to the identity
system in figure(b).

h(n) h1 (n)
x(n)
w(n)=y(n)
y(n)

(a)

x(n)
x(n) Identity system δ(n)

Hence, for continuous time LTI system, the impulse response h1(n) of the inverse
system for an LTI system with impulse response h(n) if,

h[n]*h1[n]=δ[n]

Assignment cum Tutorial Questions

Objective:
1. Which of the following impulse responses corresponds to a causal LTI system?
BL-III
a) h(n)=an b) h(n)=an u(n+1) c) h(n)=an u(n-1) d) h(n)=a-n

2. Which of the following impulse responses corresponds to stable LTI system?

BL-III
a) h(n)=an b) h(n)=an u(-n) c) h(n)=an u(n-1) d) h(n)=a-n
3. If a signal, x(n)=3-n u(n) is input to an LTI system with impulse response h(n)=u(n),
what will be the output? BL-VI
3𝑛+1 −1 3𝑛 −1 3𝑛+1 −1 3𝑛 −1
a) 𝑢(𝑛) b) 𝑢(𝑛) c) 𝑢(𝑛) d) 𝑢(𝑛)
3𝑛 2.3𝑛 2.3𝑛 3𝑛

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

4. If two LTI systems with impulse response h1(n) and h2(n) are kept in cascade, what
will be the impulse response of the combined system? Gate-13 BL-II
a) h1(n) . h2(n) b) h1(n) + h2(n) c) h1(n) * h2(n) d) h1(n) / h2(n)
5. The formula y(n)= ∑∞
𝑛=−∞ x(k) h(n − k) that gives the response y(n) of the LTI

system as the function of the input signal x(n) and the unit sample response h(n) is
known as ______________ BL-I
a) Convolution sum b) Convolution product
c) Convolution Difference d) None of the mentioned
6. The impulse response of a LTI system is h(n)={1,1,1}. What is the response of the
signal to the input x(n)={1,2,3}? BL-V
a) {1,3,6,3,1} b) {1,2,3,2,1} c) {1,3,6,5,3} d) {1,1,1,0,0}
7. An LTI system is said to be causal if and only if? BL-I
a) Impulse response is non-zero for positive values of n
b) Impulse response is zero for positive values of n
c) Impulse response is non-zero for negative values of n
d) Impulse response is zero for negative values of n
8. The result of the convolution x(-n) ∗ δ(-n-n0) is Gate-15 BL-III
a) x(n+n0) b) x(n-n0) c) x(-n+n0) d) x(-n-n0)
9. The impulse response of an LTI system can be obtained by Gate-15 BL-II
a) Differentiating the unit ramp response b) Differentiating the unit step response
c) Integrating the unit ramp response d) Integrating the unit step response
10. Two discrete time systems with impulse responses h1(n) = δ (n -1) and h2(n) = δ (n
– 2) are connected in cascade. The overall impulse response of the cascaded system is
Gate-10 BL-V
a) δ (n – 1) + δ (n – 2) b) δ (n – 4) c) δ (n – 3) d) δ (n – 1) δ (n – 2)
11. A discrete time linear shift-invariant system has an impulse response h(n) with
h(0)=1, h(1)=-1. h(2)=-2, and zero otherwise. The system is given an input sequence
x(n) with x(0)=x(2)=1, and zero otherwise. The number of nonzero samples in the
output sequence y(n), and the value of y(2) are, respectively Gate-08 BL-V

a) 5, 2 b) 6, 2 c) 6, 1 d) 5, 3

Descriptive:
1. Determine the output y(n) of a LTI system with impulse response h(n)=an u(n),
|a|<1with the input sequence x(n)=u(n). BL-V

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2. Define an LTI system and explain its properties. BL-II

3. What is the condition for stability of LTI System? Prove it. BL-V
4. Determine whether the LTI system y(n) +y(n-1)=x(n)+x(n-2) is linear, causal, time
invariant and Stable. BL-V

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT-2

Z-Transform and Discrete Fourier Series

Syllabus:
Z Transform of sequence, properties of ROC, properties of Z transform, inverse Z transform- partial fraction
method. Discrete Fourier series: Fourier series for discrete time periodic signals, Fourier Transform for discrete
time non-periodic signals, energy density spectrum, relationship of Fourier transform to Z transform, frequency
response.

2.1 Z-Transform, Properties of Z transform

Z-Transform:

Definition of Z-Transform

1. Definition
The z-transform of a discrete-time signal x(n) is defined by

X ( z) =  x ( n) z
n = −
−n

where z = re j is a complex variable. The values of z for which the sum converges define a
region in z-plane referred to as the region of convergence (ROC).

2. Notation
If x(n) has a z-transform X(z), we write

x(n) ⎯→
Z
X ( z)

3. Region of Convergence (ROC) of Z-Transform

The range of variation of z for which z-transform converges is called region of convergence
of z-transform.

Properties of ROC of Z-Transforms

• ROC of z-transform is indicated with circle in z-plane.
• ROC does not contain any poles.
• If x(n) is a finite duration causal sequence or right sided sequence, then the ROC is
entire z-plane except at z = 0.
• If x(n) is a finite duration anti-causal sequence or left sided sequence, then the ROC is
entire z-plane except at z = ∞.
• If x(n) is a infinite duration causal sequence, ROC is exterior of the circle with radius
a. i.e. |z| > a.
• If x(n) is a infinite duration anti-causal sequence, ROC is interior of the circle with
radius a. i.e. |z| < a.
Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

• If x(n) is a finite duration two sided sequence, then the ROC is entire z-plane except at
z = 0 & z = ∞.

Table 1 Common z-Transform Paris

Sequence z-Transform Region of Convergence

 (n ) 1 all z

1
 n u (n) z 
1 − z −1

1
−  n u (−n − 1) z 
1 − z −1

z −1
n n u (n) z 
(1 − z −1 ) 2

z −1
− n n u(−n − 1) z 
(1 − z −1 ) 2

1 − (cos  0 ) z −1
cos( n 0 )u (n) z 1
1 − 2(cos  0 ) z −1 + z − 2

1 − (cos  0 ) z −1
sin( n 0 )u(n) z 1
1 − 2(cos  0 ) z −1 + z − 2

4. Complex z-plane
z = Re( z ) + j Im( z ) = re j

Unit circle: z =1

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Im(z)
Unit circle

Re(z)

Properties of Z-Transform

1. Linearity
If x(n) has a z-transform X(z) with a region of convergence Rx, and if y(n) has a z-transform
Y(z) with a region of convergence Ry,

w(n) = ax(n) + by(n) ⎯→

Z
W ( z ) = aX ( z ) + bY ( z )

and the ROC of W(z) will include the intersection of Rx and Ry, that is,

Rw contains R x  R y .

2. Shifting property
If x(n) has a z-transform X(z),

x(n − n0 ) ⎯→
Z
z − n0 X ( z ) .

3. Time reversal
If x(n) has a z-transform X(z) with a region of convergence Rx that is the annulus   z   ,
the z-transform of the time-reversed sequence x(-n) is

x(−n) ⎯→
Z
X ( z −1 )

and has a region of convergence 1   z  1  , which is denoted by 1 Rx .

4. Multiplication by an exponential
If a sequence x(n) is multiplied by a complex exponential  n ,

 n x(n) ⎯→
Z
X ( −1 z ) .

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

5. Convolution theorem
If x(n) has a z-transform X(z) with a region of convergence Rx, and if h(n) has a z-transform
H(z) with a region of convergence Rh,

y(n) = x(n)  h(n) ⎯→

Z
Y ( z) = X ( z) H ( z) .

The ROC of Y(z) will include the intersection of Rx and Rh, that is,

Ry contains R x  Rh .

With x(n), y(n), and h(n) denoting the input, output, and unit-sample response, respectively,
and X(z), Y(x), and H(z) their z-transforms. The z-transform of the unit-sample response is
often referred to as the system function.

6. Conjugation
If X(z) is the z-transform of x(n), the z-transform of the complex conjugate of x(n) is

x  (n) ⎯→
Z
X  (z ) .

7. Derivative
If X(z) is the z-transform of x(n), the z-transform of n k x(n) is

dX ( z )
nx(n) ⎯→
Z
−z .
dz

8. Initial value theorem

If X(z) is the z-transform of x(n) and x(n) is equal to zero for n<0, the initial value, x(0), maybe
be found from X(z) as follows:

x(0) = lim X ( z ) .
z →

Digital Signal Processing

Unit-II
Assignment-Cum-Tutorial Questions

2.1Z – Transform, Properties of ROC, Properties of Z-Transform

Objective:

1. The Z-Transform X(z) of a discrete time signal x(n) is defined as: BL-I

2. What is the set of all values of z for which X(z) attains a finite value? BL-I
a) Radius of convergence
b) Radius of divergence
c) Feasible solution

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

d) None of the mentioned

3. What the z-transform of the finite duration signal BL-II

a) 2 + 4z + 5z2 + 7z3 + z4
b) 2 + 4z + 5z2 + 7z3 + z5
c) 2 + 4z-1 + 5z-2 + 7z-3 + z-5
d) 2z2 + 4z + 5 +7z-1 + z-3

4. Choose the ROC of the signal x(n)=δ(n-k),k>0 among given options? BL-III
a) z=0
b) z=∞
c) Entire z-plane, except at z=0
d) Entire z-plane, except at z=∞

5. Calculate the z-transform of the signal x(n)=(0.5)nu(n)? BL-II

6. Which of the following series has an ROC as mentioned below? BL-II

a) α-nu(n)
b) αnu(n)
c) α-nu(-n)
d) αnu(n)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

7. What is the z-transform of the signal x(n)= -αnu(-n-1)? BL-II

8. Find the ROC of the z-transform of the signal x(n)= anu(n)+bnu(-n-1)? BL-II
a) |a|<|z|<|b|
b) |a|>|z|>|b|
c) |a|>|z|<|b|
d) |a|<|z|>|b|

9. What is the ROC of z-transform of finite duration anti-causal sequence? BL-I

a) z=0
b) z=∞
c) Entire z-plane, except at z=0
d) Entire z-plane, except at z=∞

10. What is the ROC of z-transform of an two sided infinite sequence? BL-I
a) |z|>r1
b) |z|<r1
c) r2<|z|<r1
d) None of the mentioned

11. The z-transform of a sequence x(n) which is given as is known as:

a) Uni-lateral Z-transform
b) Bi-lateral Z-transform
c) Tri-lateral Z-transform
d) None of the mentioned BL-III

12. What is the ROC of the system function H(z) if the discrete time LTI system is BIBO
stable?
a) Entire z-plane, except at z=0
b) Entire z-plane, except at z=∞
c) Contain unit circle
d) None of the mentioned BL-II

13. The ROC of z-transform of any signal cannot contain poles. BL-I
a) True
b) False

14. Is the discrete time LTI system with impulse response h(n)=an(n) (|a| < 1) BIBO stable?

BL-III
a) True

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

b) False

15. What is the ROC of a causal infinite length sequence? BL-II

a) |z|<r1
b) |z|>r1
c) r2<|z|<r1
d) None of the mentioned

16. Which of the following justifies the linearity property of z-transform?[x(n)↔X(z)]. BL-
III
a) x(n)+y(n) ↔X(z)Y(z)
b) x(n)+y(n) ↔X(z)+Y(z)
c) x(n)y(n) ↔X(z)+Y(z)
d) x(n)y(n) ↔X(z)Y(z)

17. What is the z-transform of the signal x(n)=sin(jω0n)u(n)? BL-II

18. According to Time shifting property of z-transform, if X(z) is the z-transform of x(n) then
what is the z-transform of x(n-k)? BL-II
a) zkX(z)
b) z-kX(z)
c) X(z-k)
d) X(z+k)

19. If X(z) is the z-transform of the signal x(n) then what is the z-transform of anx(n)? BL-II
a) X(az)
b) X(az-1)
c) X(a-1z)
d) None of the mentioned
20. If the ROC of X(z) is r1<|z|<r2, then what is the ROC of X(a-1z)? BL-II

a) |a|r1<|z|<|a|r2
b) |a|r1>|z|>|a|r2
c) |a|r1<|z|>|a|r2
d) |a|r1>|z|<|a|r2

21. If X(z) is the z-transform of the signal x(n), then what is the z-transform of the signal x(-
n)?
a) X(-z)
b) X(z-1)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

c) X-1(z)
d) None of the mentioned BL-II

22. X(z) is the z-transform of the signal x(n), then what is the z-transform of the signal nx(n)?
a) -z(dX(z))/dz
b) zdX(z)/dz
c) -z-1dX(z)/dz
d) z-1(dX(z))/dz BL-II

23. What are the values of z for which the value of X(z)=0? BL-I
a) Poles
b) Zeros
c) Solutions
d) None of the mentioned
24. (BL-III)

25. (BL-III)

26. (BL-III)

Descriptive:
1. Define Z transform, What are the two types of Z transform? BL-I
2. Define unilateral Z transform BL-I
3. What is region of Convergence & Justify the Properties of ROC.
BL-II
4. Prove the following properties of Z-Transform BL-III
i. time shifting property of Z transform
ii. differentiation property in Z domain.
iii. convolution property of Z transform

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

5. Calculate the Z-Transform & Illustrate ROC for the following signals BL-III
i. X(n) =u(n)
ii. X(n)=anu(n)
iii. X(n)=10sin(0.25nπ)u(n)
iv. X(n)=e-0.1n cos(0.25nπ)u(n)

6. Given the following sequences , find Z-transform of their convolution BL-II

X1(n) =3 𝛿(n)+2 𝛿(n-1) & x2(n) =2 𝛿(n)- 𝛿(n-1)

2.2 inverse Z transform- partial fraction method.

The Inverse z-Transform

The z-transform is a useful tool in linear systems analysis. However, just as important as
techniques for finding the z-transform of a sequence are methods that may be used to invert the
z-transform and recover the sequence x(n) from X(z). Three possible approaches are described
below.

• Partial Fraction Expansion

For z-transforms that are rational functions of z,

a simple and straightforward approach to find the inverse z-transform is to perform a partial
fraction expansion of X(z). Assuming that p > q, and that all of the roots in the denominator are
simple,  i   k for i  k, X(z) may be expanded as follows:

Eq(3)

for some constants Ak for k = 1,2, . . . , p. The coefficients Ak may be found by multiplying both
sides of Eq. (3) by (1 -  k z−1) and setting z =  k . The result is

If p  q, the partial fraction expansion must include a polynomial in z−1of order (p-q). The
coefficients of this polynomial may be found by long division (i.e., by dividing the numerator
polynomial by the denominator). For multiple-order poles, the expansion must be modified.
For example, if X(z) has a second-order pole at z =  k, the expansion will include two terms,

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

where B1, and B2 are given by

2.2 Inverse Z-Transform, Partial fractions

1. Which of the following method is used to find the inverse z-transform of a signal?
a) Counter integration
b) Expansion into a series of terms
c) Partial fraction expansion
d) All of the mentioned BL-II
2. What is the inverse z-transform of X(z)=1/(1-1.5z-1+0.5z-2 ) if ROC is |z|>1?
a) {1,3/2,7/4,15/8,31/16,….}
a) {1,2/3,4/7,8/15,16/31,….}
a) {1/2,3/4,7/8,15/16,31/32,….}
d) None of the mentioned BL-II

3. What is the inverse z-transform of X(z)=1/(1-1.5z-1+0.5z-2 ) if ROC is |z| < 0.5?

a) {….62,30,14,6,2}
b) {…..62,30,14,6,2,0,0}
c) {0,0,2,6,14,30,62…..}
d) {2,6,14,30,62…..} BL-II

4. What is the inverse z-transform of X(z)=log(1+az-1) |z|>|a|?

d)None of the mentioned BL-II

5. The proper fraction and polynomial form of the improper rational transform
X(z)= (1+3z-1+11/6 z-2+1/3 z-3)/(1+5/6 z-1+1/6 z-2 )?
a) 1+2z -1+(1/6 z-1)/(1+5/6 z-1+1/6 z-2 )
b) 1-2z -1+(1/6 z-1)/(1+5/6 z-1+1/6 z-2 )
c) 1+2z -1+(1/3 z-1)/(1+5/6 z-1+1/6 z-2)
d) 1+2z -1-(1/6 z-1)/(1+5/6 z-1+1/6 z-2 ) BL-II

6. The partial fraction expansion of the proper function X(z)= 1/(1-1.5z-1+0.5z-2 )?

a) 2z/(z-1)-z/(z+0.5)
b) 2z/(z-1)+z/(z-0.5)
c) 2z/(z-1)+z/(z+0.5)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

d) 2z/(z-1)-z/(z-0.5) BL-III

7. The partial fraction expansion of X(z)= (1+z-1)/(1-z-1+0.5z-2 )?

a) (z(0.5-1.5j))/(z-0.5-0.5j) – (z(0.5+1.5j))/(z-0.5+0.5j)
b) (z(0.5-1.5j))/(z-0.5-0.5j) + (z(0.5+1.5j))/(z-0.5+0.5j)
c) (z(0.5+1.5j))/(z-0.5-0.5j) – (z(0.5-1.5j))/(z-0.5+0.5j)
d) (z(0.5+1.5j))/(z-0.5-0.5j) + (z(0.5-1.5j))/(z-0.5+0.5j) BL-II

8. The Partial fraction expansion of X(z)=1/((1+z-1 )(1-z-1)2)?

a) z/(4(z+1)) + 3z/(4(z-1)) + z/(2〖(z+1)〗2 )
b) z/(4(z+1)) + 3z/(4(z-1)) – z/(2〖(z+1)〗2 )
c) z/(4(z+1)) + 3z/(4(z-1)) + z/(2〖(z-1)〗2 )
d) z/(4(z+1)) + z/(4(z-1)) + z/(2〖(z+1)〗2 ) BL-II

9. Derive the inverse z-transform of X(z)= 1/(1-1.5z-1+0.5z2-2 ) if ROC is |z|>1?

a) (2-0.5n)u(n)
b) (2+0.5n)u(n)
c) (2n-0.5n)u(n)
d) None of the mentioned BL-II

10. The inverse z-transform of X(z)= 1/(1-1.5z-1+0.5z-2 ) if ROC is |z|<0.5? BL-II

a) [-2-0.5n]u(n)
b) [-2+0.5n]u(n)
c) [-2+0.5n]u(-n-1)
d) [-2-0.5n]u(-n-1)

11. What is the inverse z-transform of X(z)= 1/(1-1.5z-1+0.5z-2 ) if ROC is 0.5<|z|<1?

a) -2u(-n-1)+(0.5)nu(n)
b) -2u(-n-1)-(0.5)nu(n)
c) -2u(-n-1)+(0.5)nu(-n-1)
d) 2u(n)+(0.5)nu(-n-1) BL-II

Descriptive Questions:

1. Determine the causal signal x(n) having the Z-transform BL-III

1
𝑋(𝑍) =
(1 + 𝑧 )(1 − 𝑧 −1 )2
−1

2. Determine the Inverse Z-Transform of BL-II

𝑧 3 − 10𝑧 2 − 4𝑧 + 4
𝑋(𝑧) =
2𝑧 2 − 2𝑧 − 4

3. Find the Inverse Z-transform of BL-III

−1 −2
1−𝑧 +𝑧
𝑋(𝑧) =
(1 − 0.5𝑧 −1 )(1 − 2𝑧 −1 )(1 − 𝑧 −1 )

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2.3 Fourier series & Fourier Transform for DT non-periodic signals, energy density
spectrum, relationship of F.T & Z-Transform, frequency response.

DISCRETE FOURIER SERIES:

FOURIER SERIES FOR DISCRETE TIME PERIODIC SIGNALS:

A real, N-periodic, discrete-time signal x[n] can be represented by a linear combination of the
complex exponential signals

as

In these expressions, , and the discrete-time fundamental frequency is .

This discrete-time Fourier series representation provides notions of frequency content of
discrete-time signals, and it is very convenient for calculations involving linear, time-invariant
systems because complex exponentials are eigen functions of LTI systems.

The complex coefficients can be calculated from the expression

The are called the spectral coefficients of the signal x[n]. A plot of vs k is called
the magnitude spectrum of x[n], and a plot of vs k is called the phase spectrum of x[n]. These
plots, particularly the magnitude spectrum, provide a picture of the frequency composition of
the signal. Notice that the spectral coefficients repeat as k is varied. In particular, for any value
of k,

PROPERTIES:
[k] = *[-k]
[k] = [-k]
Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

∡ [k] = -∡ [-k]
Re [k] = Re [-k]
Im [k] = -Im [-k]

Finally, if [n] is real and symmetric, then the DFS is real:

while, for real antisymmetric signals, we can state that the DFS is purely imaginary.

Linearity and Shifts. The DFS is a linear operator, since it can be expressed as a matrix-vector
product. A shift in the discrete-time domain leads to multiplication by a phase term in the
frequency domain:

while multiplication of the signal by a complex exponential of frequency multiple of 2π∕N leads
to of a shift in frequency:

Energy Conservation. We have already seen the energy conservation property in the context
of basis expansion. Here, we simply recall Parseval’s theorem, which states

POWER DENSITY SPECTRUM:

Describes how power of a signal or time series is distributed over frequency, as in the simple
example given previously. Here, power can be the actual physical power, or more often, for
convenience with abstract signals, is simply identified with the squared value of the signal.
• The spectrum of a real valued process (or even a complex process using the above

definition) is real and an even function of frequency: .

• If the process is continuous and purely in deterministic, the auto covariance function can
be reconstructed by using the Inverse Fourier transform
• The PSD can be used to compute the variance (net power) of a process by integrating over
frequency

FOURIER TRANSFORM FOR DISCRETE TIME APERIODIC SIGNALS:

Discrete-time Fourier transform (DTFT) is a form of Fourier analysis that is applicable to

the uniformly-spaced samples of a continuous function. The term discrete-time refers to the
fact that the transform operates on discrete data (samples) whose interval often has units of
time. From only the samples, it produces a function of frequency that is a periodic
summation of the continuous Fourier transform of the original continuous function.

PROPERTIES:

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

ENERGY DENSITY SPECTRUM:

RELATIONSHIP OF FOURIER TRANSFORM TO Z-TRANSFORM:

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

FREQUENCY RESPONSE:
The frequency response is an extremely insightful description for linear time invariant (LTI)
systems. The output of a LTI system at a given frequency is simply the product of the input at
that frequency and the frequency response. In this lesson you will learn the relationship
between the frequency response, impulse response, differential equation, and system function
descriptions for discrete-time LTI systems. Each of these different ways of describing the
input-output behavior of LTI systems provides a unique perspective on the system
characteristics.

Properties of the Frequency Response

Existence The definition of the frequency response in terms of h[m] and sines and cosines in
Equation, or equivalently in terms of h[m] and complex exponentials in Equation, generally
involves summing an infinite number of terms, so again (just as with convolution) one needs
conditions to guarantee that the sum is well-behaved. One case, of course, is where h[m] is
nonzero at only a finite number of time instants, in which case there is no problem with the
sum. Another case is when the function h[·] is absolutely summable, ∞Σ n=−∞ |h[n]|≤μ <∞, as
this ensures that the sum defining the frequency response is itself absolutely summable.

The absolute summability of h[·] is the condition for bounded-input bounded-output (BIBO)
stability of an LTI system that we obtained in the previous chapter. It turns out that under this
condition the frequency response is actually a continuous function of Ω. Various other
important properties of the frequency response follow quickly from the definition.

Periodicity in Ω Note first that H(Ω) repeats periodically on the frequency (Ω) axis, with
Period 2π, because a sinusoidal or complex exponential input of the form in Equation or
(12.9) is unchanged when its frequency is increased by any integer multiple of 2π. This can
also be seen from Equation , the defining equation for the frequency response. It follows that
only the interval |Ω|≤π is of interest.

Lowest Frequency An input at the frequency Ω = 0 corresponds to a constant (or “DC”,

which stands for direct current, but in this context just means “constant”) input, so H(0) =
∞Σ n=−∞ h[n]

2.3DFS & DTFT, Energy Density Spectrum, Relationship of DTFT &

Z- Transforms
Objective Questions:

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

BL-I

BL-I

BL-I

BL-I

4.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

BL-II
5.

BL-II

6. DTFT of the sequences u[n-m] is

BL-II

𝑒 −𝑗𝑤𝑚 𝑒 −𝑗𝑤𝑚 𝑒 −𝑗𝑤𝑚 𝑒 𝑗𝑤𝑚

a) b) c) d)
1+𝑒 −𝑗𝑤 1−𝑒 −𝑗𝑤 1−𝑒 𝑗𝑤 1−𝑒 −𝑗𝑤

7. The frequency response of LTI system whose h[n] = [1,-2,2,3] is ______________

a) 1+2e-jw-2e-j2w-3e-j3w b) 1-2ejw-2e-j2w-3ej3w

c) 1-2e-jw+2e-j2w+3e-j3w d) 1+2e-jw+2ej2w-3e-j3w
BL-II

11. The frequency response of the system y[n] = x[n]-2x[n-1]+x[n-2] is ____________

BL-II

a) 1-2e-jw+e-j2w b) 1+2e-jw-e-j2w c) 1-2ejw-e-j2w d) 1+2ejw+ej2w

1
12. Let x (n) = (2)n u(n), y(n) = x2(n), and Y(ejw) be the Fourier transform of y(n). Then
Y(ej0) is
1 4
a) 4 b) 2 c) 4 d) 3 GATE-05 Ans: D BL-
III

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2 𝐴𝑒−𝑗6𝜋𝑓
A Fourier transform pair is given by (3)n u[n+3] −𝑗2𝜋𝑓 where u[n] denotes
1−(2/3)𝑒
the unit step sequence. The values of A is ____________ GATE-14 Ans: 3.375
BL-III
𝑛
x ( − 1) for n even
13. The sequence y (n) = { 2 GATE-05 Ans: A BL-
0 for 𝑛 𝑜𝑑𝑑
III
a) ½ 1 2 1 ½ b) ½ 1 2 1 ½

-2 0 2 4 6 n -3 -1 1 0 5 n
c) ½ 1 2 1 ½ d) ½ 1 2 1 ½

-6 -4 -2 0 2 n -5 -3 -1 1 3 n
14. The fourier transform of y(2n) will be GATE-05 Ans: C

a) e-j2w [cos4w + 2 cos2w + 2] b) [cos2w + 2 cosw + 2] (BL-III)

c) e-jw [cos2w + 2 cosw + 2] d) ej2w [cos2w + 2 cos w + 2]

15. A 5- point sequence x[n] is given as x[-3] = 1, x[-2] = 1, x[-1] = 0, x[0] = 5, x[1] =1. Let
𝜋
X(ejw) denote the discrete time Fourier transform of x[n]. The value of ∫−𝜋 X(ejw) dw is
BL-III

a) 5 b) 10π c) 16π d) 5+j10π GATE-07 Ans: B

Descriptive Questions
1. Explain the Fourier series for Discrete Time Periodic signals (DTFS). What is power
density
Spectrum of discrete time periodic signals? BL-I
2. Write notes on properties of DTFS? BL-II
5. Explain Fourier Transform for Discrete Time Aperiodic Signals? BL-II
6. State and prove the properties of DTFT? BL-III
7. Explain energy density spectrum and Frequency Response? BL-II
8. Derive the relationship between Fourier Transform and Z Transform. BL-III
3 1
9. Find frequency response of the following system y[n]-y[n-1]+16y[n-2 ]= x[n]- 2x[n-1]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

1
𝑒 𝑗𝑤[𝑒 𝑗𝑤−( )]
2
Ans : 3 BL-III
𝑒 𝑗2𝑤−𝑒 𝑗𝑤 +
16

10. Determine magnitude and phase response of the system y[n]-5y[n-1] = x[n]+4x[n-1]

√17+8𝑐𝑜𝑠𝑤
Ans : |H(w)| = BL-III
√26−10𝑐𝑜𝑠𝑤
𝑠𝑖𝑛𝑤 𝑠𝑖𝑛𝑤
∟H(w) = tan-1( )-tan-1(𝑐𝑜𝑠𝑤−5)
4+𝑐𝑜𝑠𝑤

11. Determine the signal x[n] for the Fourier Transform X(w) = e-jw for -π ≤ w ≤ π
sinπ[𝑛−1]
Ans : BL-II
π[n−1]

12. Find the exponential forms of DFS representation of x(n) shown in the figure and
comment. BL-III
3
2
x(n)

-3 -2 -1 0 1 2 3 4 5 6 7 8 9…n
Figure: x(n)

13. Evaluate the energy density spectrum of the signal x[n] = an u[n] : -1≤a<1 BL-II

14. Evaluate the DTFT of the signal BL-II

x[n] = A |n| ≤N
0 |n| >N

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT-3
DISCRETE FOURIER TRANSFORM
SYLLABUS:
Frequency sampling-Discrete Fourier Transform(DFT), Properties of DFT,
Linear Convolution of sequences using DFT, Relationship between DFT and Z
transform.

3.1Frequency sampling & DFT Properties

3.1.1 Frequency sampling-Discrete Fourier Transform (DFT)

Discrete Fourier transform is of a sequence is periodic in the frequency range of 0 to 2π. If a
digital computer used to compute N equally spaced points over the interval 0 ≤ ω< 2π, then the
N points are located at

2π
𝜔𝑘 = 𝑘 𝑘 = 0, 1, … . . 𝑁 − 1
𝑁
These N equally spaced frequency samples of the DTFT are known as DFT and denoted by
X(k) is

𝑋(𝑘) = 𝑋(𝜔)|𝜔=2π𝑘 𝑘 = 0, 1, … . . 𝑁 − 1
𝑁

2π
When the Fourier transform is sampled with sampling period the corresponding discrete-
𝑁
time sequence 𝑥𝑝 (𝑛)becomes periodic in time with period N where
∞

𝑥𝑝 (𝑛) = ∑ 𝑥(𝑛 − 𝑙𝑁)

𝑙=−∞

Thus the periodic sequence 𝑥𝑝 (𝑛), corresponding to X(k) for k = 0 to N-1 obtained by sampling
𝑋(𝜔) in the interval 0 to 2π is formed from x(n) by adding together an infinite number of
shifted replicas of x(n). This is illustrated in the figure.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Let us consider the sequence x(n) is of length L and the number of sample points as N. If N >
L then the delayed replicas of x(n) do not overlap and one period of the periodic sequence
𝑥𝑝 (𝑛) is recognizable as x(n). If N<L, then the replicas of x(n) overlap and one period of 𝑥𝑝 (𝑛)
is not identical to x(n). This is called as time domain aliasing due to under sampling of the Fourier
transform of x(n).

If N > L, x(n) can be recovered using the equation

𝑥𝑝 (𝑛) 0 ≤ 𝑛 ≤ 𝑁 − 1
𝑥(𝑛) = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Let x(n) is a causal, finite duration sequence containing L samples, then its Fourier transform
is given by
𝐿−1

𝑋(𝜔) = ∑ 𝑥(𝑛) 𝑒 −𝑗𝜔𝑛

𝑛=0

If we sample 𝑋(𝜔) at N equally spaced points over 0 ≤ ω< 2π, we obtain

𝐿−1

𝑋(𝑘) = 𝑋(𝜔)|𝜔=2π𝑘 = ∑ 𝑥(𝑛) 𝑒 −𝑗2𝜋𝑛𝑘/𝑁

𝑁
𝑛=0

To prevent time domain aliasing, the duration of x(n) has to increase from L to N by appending
N-L zeros. This is called as zero padding.

Since zero values are added the summation does not change. So

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

𝑁−1

𝑋(𝑘) = ∑ 𝑥(𝑛) 𝑒 −𝑗2𝜋𝑛𝑘/𝑁 0 ≤ 𝑘 ≤ 𝑁 − 1

𝑛=0

This is called as N-point DFT.

The inverse DFT can be obtained using

𝑁−1
1
𝑥(𝑛) = ∑ 𝑋(𝑘) 𝑒 𝑗2𝜋𝑛𝑘/𝑁 0 ≤ 𝑛 ≤ 𝑁 − 1
𝑁
𝑘=0

The twiddle factor or the Nth root of unit is defined as 𝑊𝑁 = 𝑒 −𝑗2𝜋/𝑁 then the DFT pair can
be defined as
𝑁−1

𝑋(𝑘) = ∑ 𝑥(𝑛) 𝑊𝑁𝑛𝑘 0 ≤ 𝑘 ≤ 𝑁 − 1

𝑛=0

𝑁−1
1
𝑥(𝑛) = ∑ 𝑋(𝑘) 𝑊𝑁−𝑛𝑘 0 ≤ 𝑛 ≤ 𝑁 − 1
𝑁
𝑘=0

3.1.2 Properties of the DFT

1. Periodicity
If X(k) is N-point DFT of a finite duration sequence x(n) then
𝑥(𝑛 + 𝑁) = 𝑥(𝑛) 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑛
𝑋(𝑘 + 𝑁) = 𝑋(𝑘) 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑘
2. Linearity
If 𝐷𝐹𝑇[𝑥1 (𝑛)] = 𝑋1 (𝑘) and 𝐷𝐹𝑇[𝑥2 (𝑛)] = 𝑋2 (𝑘) then
𝐷𝐹𝑇[𝑎𝑥1 (𝑛) + 𝑏𝑥2 (𝑛)] = 𝑎𝑋1 (𝑘) + 𝑏𝑋2 (𝑘)
3. Circular shift of a sequence
Circular shift of a sequence can be represented as 𝑥((𝑛 − 𝑚))𝑁 = 𝑥(𝑁 − 𝑚 + 𝑛). If
the value of N-m+n is not in between 0 and N-1, then add or subtract multiples of N
until the result is in between 0 to N-1.
If X(k) is N-point DFT of a finite duration sequence x(n) then
𝑗2𝜋𝑘𝑚
𝐷𝐹𝑇 [𝑥((𝑛 − 𝑚))𝑁 ] = 𝑒 − 𝑁 𝑋(𝑘)

4. Time reversal of the sequence

If 𝐷𝐹𝑇[𝑥(𝑛)] = 𝑋(𝑘) then

𝐷𝐹𝑇 [𝑥((−𝑛))𝑁 ] = 𝐷𝐹𝑇[𝑥(𝑁 − 𝑛)] = 𝑋((−𝑘))𝑁 = 𝑋(𝑁 − 𝑘)

5. Circular frequency shift

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

If 𝐷𝐹𝑇[𝑥(𝑛)] = 𝑋(𝑘) then

𝐷𝐹𝑇[𝑥(𝑛)𝑒 𝑗2𝜋𝑙𝑛/𝑁 ] = 𝑋((𝑘 − 𝑙))𝑁
6. Complex conjugate property
If 𝐷𝐹𝑇[𝑥(𝑛)] = 𝑋(𝑘) then
𝐷𝐹𝑇[𝑥 ∗ (𝑛)] = 𝑋 ∗ (𝑁 − 𝑘) = 𝑋 ∗ ((−𝑘))𝑁
and 𝐷𝐹𝑇[𝑥 ∗ (𝑁 − 𝑛)] = 𝑋 ∗ (𝑘)
Note: If x(n) is a real function i.e. x*(n) = x(n), then
𝑋(𝑘) = 𝑋 ∗ (𝑁 − 𝑘)

7. Circular Convolution

If 𝐷𝐹𝑇[𝑥1 (𝑛)] = 𝑋1 (𝑘) and 𝐷𝐹𝑇[𝑥2 (𝑛)] = 𝑋2 (𝑘) then

𝐷𝐹𝑇[𝑥1 (𝑛) 𝑥2 (𝑛)] = 𝑋1 (𝑘) 𝑋2 (𝑘)
where is circular convolution
8. Circular Correlation
If 𝐷𝐹𝑇[𝑥(𝑛)] = 𝑋(𝑘) and 𝐷𝐹𝑇[𝑦(𝑛)] = 𝑌(𝑘) then
𝑁−1

𝐷𝐹𝑇[𝑟̃𝑥𝑦 (𝑙)] = 𝐷𝐹𝑇 [∑ 𝑥(𝑛)𝑦 ∗ ((𝑛 − 𝑙))𝑁 ] = 𝑋(𝑘)𝑌 ∗ (𝑘)

𝑛=0

Where 𝑟̃𝑥𝑦 (𝑙) is circular cross-correlation sequence.

9. Multiplication of two sequences
If 𝐷𝐹𝑇[𝑥1 (𝑛)] = 𝑋1 (𝑘) and 𝐷𝐹𝑇[𝑥2 (𝑛)] = 𝑋2 (𝑘) then
1
𝐷𝐹𝑇[𝑥1 (𝑛)𝑥2 (𝑛)] = [𝑋 (𝑘) 𝑋2 (𝑘)]
𝑁 1
10. Parseval’s theorem
If 𝐷𝐹𝑇[𝑥(𝑛)] = 𝑋(𝑘) and 𝐷𝐹𝑇[𝑦(𝑛)] = 𝑌(𝑘) then
𝑁−1 𝑁−1
1
∑ 𝑥(𝑛)𝑦 ∗ (𝑛) = ∑ 𝑋(𝑘)𝑌 ∗ (𝑘)
𝑁
𝑛=0 𝑘=0

3.2 Linear Convolution & Relationship between DFT and Z-Transform

3.2.1 Evaluation Circular Convolution

Two methods are used to find the circular convolution of two sequences: Concentric circle
method and Matrix method.

1. Concentric circle method

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Given two sequences x1(n) and x2(n), the circular convolution of the two sequences is
x3(n) = 𝑥1 (𝑛) 𝑥2 (𝑛)can be found by using the following steps

i. Graph N samples of x1(n) as equally spaced points around first circle in counter
clockwise direction.
ii. Start at the same point as x1(n), graph N samples of x2(n) as equally spaced points
around second circle in clockwise direction.
iii. Multiply corresponding samples on the two circles and sum the products to produce
output.
iv. Rotate the second circle one sample at a time in anti-clockwise direction and go to
step (iii) to obtain the next value of output.
v. Repeat step (iv) until the second circle first sample lines up with the first sample of
the first circle once again.

x1(1) x2(3)

x1(2) x1(0) x2(2) x2(0)

Ex:Find the circular convolution of the two sequences x1(n) = {1, 2, 2, 1} and x1(n) = {1, 2, 3,
1} using concentric circle method.
x1(3) x2(1)
Sol:
2 1

2 1 3 1

1 2

x3(0)= 1(1) + 2(1) + 2(3) + 1(2) = 11

2 1

2 1 1 2

1 3
x3(1)= 1(2) + 2(1) + 2(1) + 1(3) = 9

2 2

2 1 1 3

1 1

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

x3(2)= 1(3) + 2(2) + 2(1) + 1(1) = 10

2 3

2 1 2 1

1 1

x3(3) = 1(1) + 2(3) + 2(2) + 1(1) = 12

x3(n) = {11, 9, 10, 12}

2. Matrix Method
The circular convolution can be obtained by representing the matrix form as shown in
below
𝑥2 (0) 𝑥2 (𝑁 − 1) 𝑥2 (1) 𝑥1 (0) 𝑥1 (0)
𝑥2 (1) 𝑥2 (0) ⋮ 𝑥2 (2) 𝑥1 (1) 𝑥1 (1)
𝑥2 (2) 𝑥2 (1) 𝑥2 (3) 𝑥1 (2) = 𝑥1 (2)
⋮ ⋮ ⋮
[𝑥2 (𝑁 − 1) 𝑥2 (𝑁 − 2) 𝑥2 (0)] [𝑥1 (𝑁 − 1)] [𝑥1 (𝑁 − 1)]
The sequence x2(n) is repeated via circular shift of samples represented in N x N matrix
form. The sequence x1(n) is represented as column matrix. The multiplication of these
two matrices gives the sequence x3(n).

Ex: Find the circular convolution of the two sequences x1(n) = {1, 2, 2, 1} and x1(n) = {1, 2,
3, 1} using matrix method.

Sol: Represent x2(n) in N x N matrix form and x1(n) in column matrix form.

1132 1 11
2113 2 9
[ ][ ] = [ ]
3211 2 10
1321 1 12
x3(n) = {11, 9, 10, 12}

3.2.2 Linear Convolution of sequences using DFT (Circular Convolution)

In signal processing applications, linear convolution is mostly used as one signal is to be filtered
and other signal will be the impulse response of the system. By multiplying the DFTs of both
signals and taking inverse DFT is equivalent to circular convolution. In order to obtain the
linear convolution from circular convolution, some modifications are needed.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Let us consider two finite duration sequences x(n) and h(n) with the duration of L samples and
M samples respectively. The linear convolution of x(n) and h(n) is given by
∞

𝑦(𝑛) = ∑ 𝑥(𝑘)ℎ(𝑛 − 𝑘)
𝑘=−∞

y(n) is a finite duration sequence of L+M-1 samples.

The circular convolution of x(n) and h(n) give N samples where N = Max (L, M). To find the
circular convolution, the lengths of x(n) and h(n) make same by appending corresponding
zeros. If M < L, then L-M zeros has to be added to the sequence h(n). The circular convolution
length obtained is L which is M-1 points shorter than the length of the linear convolution.

In order to obtain the length of circular convolution as L+M-1, both x(n) and h(n) must be of
same length (L+M-1). By increasing the lengths of sequences x(n) and h(n) by adding zeros to
L+M-1 and the circular convolution of both will give the same result as linear convolution.

Ex: Determine the output response y(n) if h(n) = {1, 1, 1} and x(n) = {1, 2, 3, 1} by using
(i) Linear convolution (ii) Circular convolution (iii) Circular convolution with zero padding.

Sol:

(i) Given x(n) = {1, 2, 3, 1}, h(n) = {1, 1, 1}

here L = 4, M = 3
h(n)
* 1 1 1
1 1 1 1
2 2 2 2
x(n)
3 3 3 3
1 1 1 1
y(n) = {1, 3, 6, 6, 4, 1}
Number of samples in linear convolution is L+M-1 = 4 + 3 – 1 = 6

(ii) Given x(n) = {1, 2, 3, 1}, h(n) = {1, 1, 1}

here L = 4, M = 3. Using matrix method

1011 1 5
1101 2 4
[ ][ ] = [ ]
1110 3 6
0111 1 6
y(n) = {5, 4, 6, 6}

(iii) Given x(n) = {1, 2, 3, 1}, h(n) = {1, 1, 1}here L = 4, M = 3

To get the result of linear convolution with circular convolution, add M – 1 = 2 zeros
to x(n) and L – 1 = 3 zeros to h(n). Then
x(n) = {1, 2, 3, 1, 0, 0} and h(n) = {1, 1, 1, 0, 0, 0}. Using matrix method

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

10 00 1 1 1 1
11 00 0 1 2 3
11 10 0 0 3 = 6
0 1 1 1 0 0 1 6
0 0 1 1 1 0 0 4
[0 0 ]
0 1 1 1 0[ ] [ 1]
y(n) = {1 3, 6, 6, 4, 1}

3.2.3 Relationship between DFT and Z-Transform

Let us consider a sequence x(n) of finite duration N with z-transform
𝑁−1

𝑋(𝑧) = ∑ 𝑥(𝑛)𝑧 −𝑛
𝑛=0

1
Substitute the IDFT of X(k) i.e. 𝑥(𝑛) = 𝑁 ∑𝑁−1
𝑘=0 𝑋(𝑘) 𝑒
𝑗2𝜋𝑛𝑘/𝑁
in above equation

𝑁−1 𝑁−1
1 𝑗2𝜋𝑛𝑘
𝑋(𝑧) = ∑ [ ∑ 𝑋(𝑘) 𝑒 𝑁 ] 𝑧 −𝑛
𝑁
𝑛=0 𝑘=0

𝑁−1 𝑁−1
1 𝑛
= ∑ 𝑋(𝑘) ∑(𝑒 𝑗2𝜋𝑘/𝑁 𝑧 −1 )
𝑁
𝑘=0 𝑛=0

𝑁−1
1 − 𝑎𝑁
𝑊𝑒 𝑘𝑛𝑜𝑤 𝑡ℎ𝑎𝑡 ∑ 𝑎𝑛 =
1−𝑎
𝑛=0

𝑁−1
1 1 − (𝑒 𝑗2𝜋𝑘/𝑁 𝑧 −1 )𝑁
𝑋(𝑧) = ∑ 𝑋(𝑘)
𝑁 1 − 𝑒 𝑗2𝜋𝑘/𝑁 𝑧 −1
𝑘=0

𝑁−1
1 − 𝑧 −𝑁 𝑋(𝑘)
𝑋(𝑧) = ∑
𝑁 1 − 𝑒 𝑗2𝜋𝑘/𝑁 𝑧 −1
𝑘=0

𝑆𝑖𝑛𝑐𝑒 𝑒 𝑗2𝜋𝑘 = 1 𝑘 = 0, 1, …. , 𝑁 − 1

Assignment cum Tutorial Questions

3.1 Frequency sampling & DFT Properties

Objective:
1. The DFT of a time reversed sequence [x(-n),mod N ] is ___________ BL-II

a) X[(k) mod N] b) X[N mod (-k)] c) X[(-k) mod N] d) X[N mod (k)]
2. DFT of Circular convolution of sequences x[n] and y[n] BL-II

a) X(k) Y*(k) b) X(k) Y(k) c) X(k)*Y(k) d) X*(k) Y(k)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

3. If DFT of x[n] is X[k], then DFT of x(n-m) is ______________ BL-II

a) X(-k)e(j2πmk)/N b) X(k)ej2πnk/N c) X(-k)e(-j4πnk)/N d) X(k)e(-j2πmk)/N

4.Match the following BL-III
I. WNk+N/2 A. WNk
II. WNk+N B. -WNk
III. WN C. e-j2π/N
IV. WNWN* D. NIN
a) I equals to A, II equals to B, III equals to C, IV equals to D
b) I equals to D, II equals to C, III equals to B, IV equals to A
c) I equals to B, II equals to A, III equals to C, IV equals to D

d) I equals to C, II equals to A, III equals to D, IV equals to B

5. Find the Magnitude and phase angle of W84 BL-III

a) 1, -π b) -π, 1 c) π, -π d)-1, 1
6. If x[n] = [4, 3, 2, 1] then x((n-2))4 = BL-III

a) [4, 3, 2, 1] b) [1, 2, 3, 4] c) [2, 1, 4, 3] d) [3, 4, 1, 2]

7. The 4-point DFT of sequences x[n] = [0, 2, 4, 6] is BL-III

a){1, 1+j2, 8, 8-j2} b){21, -2+j3, -1, -5-j}

c){6, 4+j8, -1, -8+j} d) {12, -4+j4, -4, -4-j4}

8.N-point DFT of x[n] = δ[n] is _______________ BL-III
a) X(k) = 1 for 0 ≤ k ≤ N-1 b) X(k) = 1 for 0 ≤ k ≤ N+1

c) X(k) = 2 for 0 ≤ k ≤ N+1 d) X(k) = 1 for N+1 ≤ k ≤ N-1

9. The IDFT of x[k] = [1, 0, 1, 0] is __________________ BL-III

a) [0, 0.5, 0, 0.5] b) [0, 1, 0, 1] c) [0.5, 0, 0.5, 0] d) [0, 1, 0.5, 0]

10. If x(n) is a real-valued periodic sequence with a period N. x(n) and X(k) form N-point
1
Discrete Fourier Transform (DFT) pairs. The DFT of y(n)= 𝑁 ∑𝑁−1
𝑟=0 𝑥(𝑟)𝑥(𝑛 + 𝑟) is
(GATE-08)BL-V
1 1
a) |X(k)|2 b) 𝑁 ∑𝑁−1
𝑟=0 𝑋(𝑟)𝑋 ∗ (𝑘 + 𝑟) c) ∑𝑁−1
𝑟=0 𝑋(𝑟)𝑋(𝑘 + 𝑟) d) 0
𝑁

11. The 4-point Discrete Fourier Transform (DFT) of a discrete time sequence {1, 0, 2, 3} is
(GATE-09) BL-III

a) [0, -2+2j, 2, 2-2j] b) [2, 2+2j, 6, 2-2j]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

c) [6, 1-3j, 2, 1+3j] d) [6, -1+3j, 0, -1-3j]

1 2𝜋 nk
12. The N-point DFT of sequence x[n], 0≤ n ≤ N-1is given by X[k] = ∑𝑁−1
𝑛=0 𝑥[𝑛]e
-j
√𝑁 𝑁

0 ≤ k ≤ N-1 Denote this relation as X=DFT(x). For N=4 which one of the following
sequence satisfies DFT (DFT(x)) = _____ (GATE-14)BL-V

a) x = [1 2 3 4] b) [1 2 3 2] c) x=[1 3 2 2] d) x=[1 2 2 3]
13.The first six point of the 8-points DFT of a real valued sequence are 5,1-j3,0,3-j4,0 and
3+j4.The last two points of the DFT are respectively. (GATE – 11) BL-V

a) 0, 1-j3 b) 0, 1+j3 c) 1+j3, 5 d) 1-j3, 5

Descriptive:
1. Explain Discrete Fourier transform and Inverse Discrete Fourier transform through
frequency sampling concept. BL-II
2. State and prove any three properties of DFT. BL-II
3. Find DFT of sequence
1 0 ≤ n ≤ 2
𝑥(𝑛) {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
For N = 8 and plot │X(k)│ and ∟X(k) BL-III

4. Find IDFT of sequence X(k)= {5, 0, 1-j, 0, 1, 0, 1+j, 0} BL-III

5. Find N-point DFT of the following sequence x(n) = an 0 ≤ n ≤ N-1 BL-III
6. If the DFT of sequence x(n) = {1, 2, 2, 1} is X(k). Plot the sequence whose DFT is
Y(k) = e-(4πk)/5 X(k) BL-IV

3.2 Linear Convolution & Relationship between DFT and Z-Transform

Objective:
1. Circular convolution of x1(n) = {1, 1, 2, 1} and x2(n) = {1, 2, 3, 4} is __________ BL-II

a){11, 12, 13, 14} b) {13, 14, 11, 12} c){3, 4, 5, 6} d) {1, 14, 15, 12}

2. If the length of x(n) is L and length of h(n) is M, then what will be the length of both

sequences so that circular convolution will give same sequence as linear convolution BL-I

a) L-1 b) L+M-1 c)M-1 d)Max (L, M)

3. Linear convolution of sequences x(n) and h(n) using DFT: x(n) = {1, 2, 3, 4},

h(n) = {1, 1, 2, 2} is BL-III

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

a){11, 12, 13, 14} b) {13, 14, 11, 12} c) {15, 17, 15, 13} d) {15, 14, 15, 12}
4. Consider two real sequences with time – origin marked by the bold value,x1(n) =
{1,2,3,0}, x2(n) ={1,3,2,1}. Let X1(k) and X2(k) be 4-point DFTs of x1(n) and x2(n),
respectively .Another sequence x3(n) is derived by taking 4-point inverse DFT of X3(k)
=X1(k)X2(k).The value of x3(2) is_____. (GATE – 15) BL-V

a) 9 b) 8 c) 11 d) 14

Descriptive:
1. Determine the output response y(n) if x(n) = {1, 2, 3, 1} and h(n) = {1, 1, 1} by using
a) Linear Convolution b) Circular Convolution c) Circular Convolution with zero padding
BL-III
2. Distinguish between linear and circular convolution of two sequences.BL-II
3. Find Z-Transform of the sequence x(n) = u(n) – u(n-8) and sample it at 6 points on the unit
circle using the relation X(k) = X(z)│z = ej2πk/6, k = 0, 1, 2, 3, 4, 5. Find the Inverse DFT of
X(k) and compare it with x[n] and give your comments. BL-V
4. Find the circular convolution of two finite duration sequences x(n) = {1, -1, -2, 3, -1} and
h(n) = {1, 2, 3} BL-III
5. Find the circular convolution of two sequences given with N = 5.
x(n) = δ(n) - δ(n-2) + δ(n-4)
h(n) = δ(n) + δ(n-1)- δ(n-2) - δ(n-3) BL-III
6. Let X(k) denote the N-point DFT of an N-point sequence x(n). If the DFT of X(k) is
computed to obtain x1(n). Determine x1(n) in terms of x(n). BL-V

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT-IV: Fast Fourier Transform

Objective: To develop FFT based DSP algorithms which are computationally efficient for
evaluating the DFT.

Syllabus: Fast Fourier transform-Radix-2 decimation in time and in frequency FFT

algorithms, IDFT using FFT algorithms, FFT for composite N.

Outcomes:

Students will be able to

➢ compute DFT by using FFT algorithms.

➢ to develop FFT flow-graphs
➢ differentiate DIT and DIF algorithms.

There are several methods for computing the DFT efficiently. In view of the importance of the
DFT in various digital signal processing applications, such as linear filtering, correlation
analysis, and spectrum analysis, its efficient computation is a topic that has received
considerable attention by many mathematicians, engineers, and applied scientists.

From this point, we change the notation that X(k), instead of y(k) in previous sections,
represents the Fourier coefficients of x(n).

Basically, the computational problem for the DFT is to compute the sequence {X(k)}
of N complex-valued numbers given another sequence of data {x(n)} of length N, according to
the formula

In general, the data sequence x(n) is also assumed to be complex valued. Similarly, The IDFT
becomes

Since DFT and IDFT involve basically the same type of computations, our discussion of
efficient computational algorithms for the DFT applies as well to the efficient computation of
the IDFT.

We observe that for each value of k, direct computation of X(k) involves N complex
multiplications (4N real multiplications) and N-1 complex additions (4N-2 real additions).
Consequently, to compute all N values of the DFT requires N 2 complex multiplications
and N 2-N complex additions.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Direct computation of the DFT is basically inefficient primarily because it does not exploit
the symmetry and periodicity properties of the phase factor WN. In particular, these two
properties are :

The computationally efficient algorithms described in this sectio, known collectively

as fast Fourier transform (FFT) algorithms, exploit these two basic properties of the
phase factor.

Radix-2 FFT Algorithms

Let us consider the computation of the N = 2v point DFT by the divide-and conquer
approach. We split the N-point data sequence into two N/2-point data
sequences f1(n) and f2(n), corresponding to the even-numbered and odd-numbered
samples of x(n), respectively, that is,

Thus f1(n) and f2(n) are obtained by decimating x(n) by a factor of 2, and hence the
resulting FFT algorithm is called a decimation-in-time algorithm.

Now the N-point DFT can be expressed in terms of the DFT's of the decimated
sequences as follows:

But WN2 = WN/2. With this substitution, the equation can be expressed as

where F1(k) and F2(k) are the N/2-point DFTs of the sequences f1(m) and f2(m),
respectively.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Since F1(k) and F2(k) are periodic, with period N/2, we have F1(k+N/2) = F1(k)
and F2(k+N/2) = F2(k). In addition, the factor WNk+N/2 = -WNk. Hence the equation
may be expressed as

We observe that the direct computation of F1(k) requires (N/2)2 complex

multiplications. The same applies to the computation of F2(k). Furthermore, there
are N/2 additional complex multiplications required to compute WNkF2(k). Hence the
computation of X(k) requires 2(N/2)2 + N/2 = N 2/2 + N/2 complex multiplications.
This first step results in a reduction of the number of multiplications from N 2 to N 2/2
+ N/2, which is about a factor of 2 for N large.

Figure1. First step in the decimation-in-time algorithm.

By computing N/4-point DFTs, we would obtain the N/2-point DFTs F1(k)

and F2(k) from the relations

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The decimation of the data sequence can be repeated again and again until the
resulting sequences are reduced to one-point sequences. For N = 2v, this decimation
can be performed v = log2N times. Thus the total number of complex multiplications
is reduced to (N/2)log2N. The number of complex additions is Nlog2N.

For illustrative purposes, Figure.2 depicts the computation of N = 8 point DFT. We

observe that the computation is performed in tree stages, beginning with the
computations of four two-point DFTs, then two four-point DFTs, and finally, one
eight-point DFT. The combination for the smaller DFTs to form the larger DFT is
illustrated in Figure.3 for N = 8.

Figure2. Three stages in the computation of an N = 8-point DFT.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure3. Eight-point decimation-in-time FFT algorithm.

Figure4. Basic butterfly computation in the decimation-in-time FFT algorithm.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure5. Shuffling of the data and bit reversal.

An important observation is concerned with the order of the input data sequence
after it is decimated (v-1) times. For example, if we consider the case where N = 8,
we know that the first decimation yeilds the sequence x(0), x(2), x(4), x(6), x(1), x(3),
x(5), x(7), and the second decimation results in the sequence x(0), x(4), x(2), x(6),
x(1), x(5), x(3), x(7). This shuffling of the input data sequence has a well-defined
order as can be ascertained from observing Figure.5, which illustrates the decimation
of the eight-point sequence.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Another important radix-2 FFT algorithm, called the decimation-in-frequency

algorithm, is obtained by using the divide-and-conquer approach. To derive the
algorithm, we begin by splitting the DFT formula into two summations, one of which
involves the sum over the first N/2 data points and the second sum involves the last
N/2 data points. Thus we obtain

Now, let us split (decimate) X(k) into the even- and odd-numbered samples. Thus
we obtain

where we have used the fact that WN2 = WN/2

The computational procedure above can be repeated through decimation of the N/2-point
DFTs X(2k) and X(2k+1). The entire process involves v = log2N stages of decimation, where
each stage involves N/2 butterflies of the type shown in Figure.7. Consequently, the
computation of the N-point DFT via the decimation-in-frequency FFT requires (N/2)log2N
complex multiplications and Nlog2N complex additions, just as in the decimation-in-time
algorithm. For illustrative purposes, the eight-point decimation-in-frequency algorithm is given
in Figure.8.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure6. First stage of the decimation-in-frequency FFT algorithm.

Figure7. Basic butterfly computation in the decimation-in-frequency

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure8. N = 8-piont decimation-in-frequency FFT algorithm.

We observe from Figure.8 that the input data x(n) occurs in natural order, but the
output DFT occurs in bit-reversed order. We also note that the computations are
performed in place. However, it is possible to reconfigure the decimation-in-
frequency algorithm so that the input sequence occurs in bit-reversed order while the
output DFT occurs in normal order. Furthermore, if we abandon the requirement that
the computations be done in place, it is also possible to have both the input data and
the output DFT in normal order.

FFT ALGORITHMS FOR N A COMPOSITE NUMBER

Till now we have discussed DIT and DIF FFT algorithm for the important special case of N a
power of 2,N=2𝑚 .They are called radix-2 FFTs. When N is a power of 2, the decomposition
leads to a highly efficient computational algorithm. Further more, all the required computations
are butterfly computations that correspond essentially to two-point DFTs. For this reason, the

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

power-of -2 algorithms are particularly simple to implement. So in some cases, even if N is not
a power of 2, it is made a power of 2 by simply augmenting with zeros. However, in some
cases it may not be possible to choose N to be a power of 2. So we have to consider composite
radix FFT .

A composite or mixed radix FFT is used when N is a composite number which has
more than one prime factor ; for example N=6 or 10 or 12.For these cases also ,efficient DIT
and DIF algorithms can be developed. Let us consider DIF FFT decomposition for N a
composite number.

If N=p1p2………pm=p1N1

Where N1=p2p3…pm,the input sequence x(n) can be separated into

p1subsequences of N1 samples each.Then the DFT can be written as

X(k) =∑𝑵−𝟏 𝒏𝒌
𝒏=𝟎 𝒙(𝒏) 𝑾𝑵

(𝒏𝒑𝟏+𝟏)𝒌
=∑𝑵𝟏−𝟏 𝑵𝟏−𝟏
𝒏=𝟎 𝒙(𝒏𝒑𝟏) + ∑𝒏=𝟎 𝒙(𝒏𝒑𝟏 + 𝟏) 𝑾𝑵 +…………..
(𝒏𝒑𝟏+𝒑𝟏−𝟏)𝒌
+∑𝑵𝟏−𝟏
𝒏=𝟎 𝒙(𝒏𝒑𝟏 + 𝒑𝟏 − 𝟏) 𝑾𝑵

Digital Signal Processing -Unit-IV

Assignment-Cum-Tutorial Questions
A. Questions testing the remembering/understanding level of students

Module-1: DITFFT & DIFFFT

I. Objective Questions

1. The cooley - tuckey algorithm of FFT is a

a) Divide and Conquer algorithm. b) Divide and Rule algorithm.
c) Split and Rule algorithm. d) Split and Combine algorithm.

2. FFT may be used to calculate 1) DFT 2) IDFT

a) true. b) false

3. DIT Algorithm divides the sequence into

a) Positive and Negative values. b) Even and Odd samples.
c) Upper higher and lower spectrum. d) Small and large samples.

4. The computational procedure for Decimation in frequency algorithm takes

a) log2N stages. b) 2log2N stages. c) log2N2 stages. d) log2N/2 stages.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

5. Which of the following is true regarding the number of computations required to compute
an N-point DFT?
a) N2 complex multiplications and N(N-1) complex additions
b) N2 complex additions and N(N-1) complex multiplications
c) N2 complex multiplications and N(N+1) complex additions
d) N2 complex additions and N(N+1) complex multiplications
Answer: a

6. Which of the following is true regarding the number of computations required to compute
DFT at any one value of ‘k’?
a) 4N-2 real multiplications and 4N real additions
b) 4N real multiplications and 4N-4 real additions
c) 4N-2 real multiplications and 4N+2 real additions
d) 4N real multiplications and 4N-2 real additions
Answer: d

7. WNk+N/2=
a) WNk
b) -WNk
c) WN-k
d) None of the mentioned
Answer: b

8. If we split the N point data sequence into two N/2 point data sequences f1(n) and f2(n)
corresponding to the even numbered and odd numbered samples of x(n), then such an FFT
algorithm is known as decimation-in-time algorithm.
a) True
b) False
Answer: a

9. If we split the N point data sequence into two N/2 point data sequences f1(n) and f2(n)
corresponding to the even numbered and odd numbered samples of x(n) and F1(k) and F2(k)
are the N/2 point DFTs of f1(k) and f2(k) respectively, then what is the N/2 point DFT X(k) of
x(n)?
a) F1(k)+F2(k)
b) F1(k)- WNk F2(k)
c) F1(k)+WNk F2(k)
d) None of the mentioned
Answer: c

10. If X(k) is the N/2 point DFT of the sequence x(n), then what is the value of X(k+N/2)?
a) X(K)
b) -X(K)
C) X(K+N)
Answer: A

11. The total number of complex multiplications required to compute N point DFT by radix-2
FFT is:
a) (N/2)log2N
b) Nlog2N

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

c) (N/2)logN
d) None of the mentioned
Answer: a

12. The total number of complex additions required to compute N point DFT by radix-2 FFT
is:
a) (N/2)log2N
b) Nlog2N
c) (N/2)logN
d) None of the mentioned
Answer: b

13. The following butterfly diagram is used in the computation of:

a) Decimation-in-time FFT
b) Decimation-in-frequency FFT
c) All of the mentioned
d) None of the mentioned
Answer: a

14. For a decimation-in-time FFT algorithm, which of the following is true?

a) Both input and output are in order
b) Both input and output are shuffled
c) Input is shuffled and output is in order
d) Input is in order and output is shuffled
Answer: c

15. In the Butterfly flow graph of radix-2 16 point DIF-FFT, the number of butterflies in each
stage is

a) 8 b) 4 c) 16 d) 2

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

16. The following butterfly diagram is used in the computation of:

a) Decimation-in-time FFT
b) Decimation-in-frequency FFT
c) All of the mentioned
d) None of the mentioned
Answer: b

17. For a decimation-in-Frequency FFT(DIF FFT) algorithm, which of the following is true?
a) Both input and output are in order
b) Both input and output are shuffled
c) Input is shuffled and output is in order
d) Input is in order and output is shuffled
Answer: d

18. FFT algorithm is designed to perform complex operations.

a) True
b) False
Answer: a

19. Decimation-in frequency FFT algorithm is used to compute DFT

a) True
b) False
Answer: a

20. For an N-point FFT algorithm with N=2m which one of the following statement is TRUE?
(GATE-10)
a) It is possible to construct a signal flow graph with both input and output in normal order
b) The number of butterflies in the mth stage is N/m
Ans: a

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

c) In-place computations requires storage of only 2N node data

d) Computation of a butterfly requires only one complex multiplication
21. The first six points of the 8-point DFT of a real valued sequence are 5, 1-j3, 0, 3-j4, 0 and
3+j4
the last two points of the DFT are respectively. (GATE-11)
a) 0, 1-j3 b) 0, 1+j3 c) 1+j3, 5 d) 1-j3, 5
Ans: b
22. The DFT of a vector [a b c d] is the vector [α β γ δ]. Consider the product (GATE-13)
𝑎 𝑏 𝑐 𝑑
𝑑 𝑎 𝑏 𝑐
[p q r s]= [a b c d][ ] The DFT of the vector [p q r s] is a scaled version of
𝑐 𝑑 𝑎 𝑏
𝑏 𝑐 𝑑 𝑎

a) [α2 β2 γ2 δ2] b) √α √ β √γ √δ c) [α+β β+γ γ+δ γ+α] d) [α β γ δ]

Ans: a
II. Descriptive Questions
1. Draw the basic butterfly diagram for DIT algorithm and DIF algorithm.
2. Derive the decimation in-time Radix-2 FFT algorithm.
3. Explain in detail about Decimation-In-Frequency FFT algorithm.
4. Compare and contrast between DIT and DIF-FFT algorithm.
5. Explain the terms a) In place computation b) Bit Reversal
6. Find the DFT of a sequence x (n) = {1, 2, 3, 4, 4, 3, 2, 1} using DIT-FFT algorithm.
7. Compute the eight-point DFT of two sequences

x (n) = 1 0 ≤ n ≤ 7
0 otherwise by using DIF-FFT algorithm

8. a) Draw 4 point radix-2 DIT-FFT butterfly structure for DFT

b) Draw 4 point radix-2 DIF-FFT butterfly structure for DFT

9. Explain FFT for composite ‘N’.

MODULE-2: IDFT using FFT algorithm

1. IDFT can be calculated using DITFFT/DIFFFT Algorithm

A) True b) False
2. Which of the following are the true for Finding IDFT using radix-2 DITFFT?

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

A) 1 only b) 1 & 2 only c) 1,2 & 3 only d) all of the above

3. The following expression is used for finding IDFT

A) True B) False
4. If X(K)={24, -j2,0, j2} then IDFT using FFT is
a) {6,7,6,5} b) {1.6.4.7} c) {1,2,3,4} d) { 4. 4. 3.2}
5. The following 4-point DITFFT Butterfly diagram is used for IDFT of
X(k)={X(0),X(1),X(3),X(4)}
Diagram:

a) True b) False

Descriptive Questions:

1. Explain how you can find IDFT using FFT algorithm.

2. Compute the IDFT of the sequences X (k) = {7, -0.707-j0.707, -j, 0.707-j0.707, 1,

0.707+j0.707, j, -0.707+j0.707 } using DIT algorithm.

3. Find the IDFT of the sequence X (k) = {4, 1-j2.414, 0, 1-j0.414, 1+j0.414, 0, 1+j2.414}

using DIF algorithm.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT-V:Design of IIR Filters

Objective:To introduce the concept of IIR Filters

Syllabus:

Analog filter approximation-Butterworth and Chebyshev (Type-1) filters, Design of IIR Filters from
analog filters- Impulse invariant technique, bilinear transformation.

Outcomes:
At the end of the Course, Student will be able todesign Analog and Digital Butterworth and
Chebyshev filters

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

DESIGN OF IIR FILTERS

Analog filter approximation-Butterworth and Chebyshev (Type-1) filters
5.1. Butterworth Filter:
Low pass Butterworth filters are all - pole filters with monotonic frequency response in both
pass band and stop band, characterized by the magnitude - squared frequency response

Where, N is the order of

the filter, Ώc is the -3dB frequency, i.e., cut-off frequency, Ώp is the pass band edge frequency
and 1= (1 /1+ε2 ) is the band edge value of │Ha(Ώ)│2. Since the product Ha(s) Ha(-s) and
evaluated at s = jΏ is simply equal to │Ha(Ώ)│2, it follows that

The poles of Ha(s)Ha(-s) occur on a circle of radius Ώc at equally spaced points. We find the
pole positions as the solution of

And hence, the N poles in the left half of the s-plane are

Note that, there are no poles on the imaginary axis of s-plane, and for N odd there will be a
pole on real axis of s-plane, for N even there are no poles even on real axis of s-plane. Also
note that all the poles are having conjugate symmetry. Thus the design methodology to design
a Butterworth low pass filter with δ2 attenuation at a specified frequency Ώs is Find N,

Where by definition, δ2 = 1/√1+δ2. Thus the Butterworth filter is completely characterized by the
parameters N, δ2, ε and the ratio Ώs/Ώp or Ώc.Then, from above Eq. find the pole positions Sk; k = 0,1,
2,……..(N-1). Finally the analog filter is given by

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure 5.1 Butterworth approximation of magnitude response.

Steps to designing analog Butterworth filter

5.2 Chebyshev Filter:

There are two types of Chebyshev filters. Type I Chebyshev filters are all-pole filters that
exhibit equiripple behavior in the pass band
and a monotonic characteristic in the stop
band. On the other hand, type II
Chebyshev filters contain both poles and zeros and exhibit a monotonic behavior in the pass
band and an equiripple behavior in the stop band. The zeros of this class of filters lie on the
imaginary axis in the s-plane. The magnitude squared of the frequency response characteristic
of type I Chebyshev filter is given as

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Where ε is a parameter of the filter related to the ripple in the pass band and TN is the Nth
order Chebyshev polynomial defined as

The Chebyshev polynomials can be generated by the recursive equation

Where T0(x) = 1 and T1(x) = x. At the band edge frequency Ώ= Ώp, we have

Or equivalently

Where δ1 is the value of the pass band ripple.

Figure 5.2: Type I Chebyshev filter characteristic

The poles of Type I Chebyshev filter lie on an ellipse in the s-plane with major axis
And minor axis

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Where β is related to ε according to the equation

The angular positions of the left half s-plane poles are given by

The poles of a chebyshev filter is obtained by

Where

Then the positions of the left half s-plane poles are given by

Where σk = r2 Cos φk and Ώk = r1Sinφk. The order of the filter is obtained from

Where, by definition δ2 = 1/√1+δ2.

Finally, the Type I Chebyshev filter is given by

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure 5.3 Chebyshev filter magnitude response

Steps to design analog Chebyshev filter:

1. From the give specifications find the order of the filter N.
2. Round off it to the next higher integer.
3. Using the following formulas find the values of a and b, which are minor and major axis of
the ellipse respectively.
4. Calculate the poles of chebyshev filter which lie on an ellipse using formula sk.
5. Find the denominator polynomial of the transfer function using the above poles.

5.3 Design of IIR filters from analog filters

There are several methods that can be used to design digital filters having an infinite duration
unit sample response. The techniques described all are based on converting an analog filter into
a digital filter and have some properties
1. The j𝛺-axis in the s-plane should map into the unit circle in the z-plane. Thus there will be
a direct relationship between the two frequency variables in the two domains.
2. The left-half of the s-plane should map into the inside of the unit circle in the z-plane. Thus
a stable analog filter will be converted into a stable digital filter.
The four most used techniques are

5.3.1. Design of IIR filter using Impulse Invariance technique

It is designed such that unit impulse response h(n) of digital filter is the sampled version of the Impulse
response of analog filter. The z-transform of an infinite impulse response is

given by

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Let us consider the mapping points from s-plane to the z-plane implied by the relation

Z=esT

If we substitute s = σ +jΩ and express the complex variable z in polar form as z = rejw then we get

r = eσTand w = ΩT

The first term in the product is eσThas a magnitude of eσT and an angle of 0 - a real number. The second
term ejΩT has unity magnitude and an angle of ΩT. Therefore, our analog pole is mapped to a place in
the z-plane of magnitude eσT and an angle ΩT.

Consider any pole on the jΩ-axis, where σ = 0 as shown in the figure. these poles map to the z-plane
at radius r = e0.T = 1. Therefore, the impulse invariant mapping map poles from the s-plane’s jΩ-axis to
the z-plane unit circle.

Figure 5.5jΩ-axis mapping to the unit circle

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure 5.6 Stable poles mapping inside the unit circle

Let Ha(s) be the system function of analog filter, expressed as
Ha(s) = ∑𝑁
𝑘=1 𝑐 k /(s-pk)

Next compute the z-transform of the digital filter, expressed as

H(z) = ∑𝑁
𝑘=1 𝑐 k /(1-e
pkT -1
z )
The final equation is given by
H(z) = ∑𝑁
𝑘=1 𝑇𝑐 k /(1-e
pkT -1
z )

Steps to design a digital filter using Impulse Invariance method:

1. For the given specifications, find Ha(s), the transverse function of an analog filter.
2. Select the sampling rate of the digital filter, T seconds per sample.
3. Express the analog filter transverse function as the sum of single pole filters.
Ha(s) = ∑𝑁
𝑘=1 𝑐 k /(s-pk)

4. Compute the z-transform of the digital filter by using the formula

H(z) = ∑𝑁
𝑘=1 𝑐 k /(1-e
pkT -1
z )
For high sampling rates use
H(z) = ∑𝑁
𝑘=1 𝑇𝑐 k /(1-e
pkT -1
z )

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

5.3.2. Deign of IIR filter using Bilinear Transformation

The bilinear transformation is a conformal mapping that transforms the jΩ-axis into the unit
circle in the z-plane only once, thus avoiding aliasing of frequency components.
Consider an analog linear filter with system function
H(s) = b/(s+a)
This can be written as
𝑌(𝑠) 𝑏
= 𝑠+𝑎 so sY(s) +aY(s) = bx(t)
𝑋(𝑠)

This can be characterized by differential equation

𝑑𝑦(𝑡)
+ay(t) = bx(t)
𝑑𝑡

y(t) can be approximated by trapezoidal formula.

𝑡
Thus y(t) = ∫𝑡0 𝑦 ′ (𝑇)𝑑𝑇 + 𝑦(𝑡0)
Where t = nT and t0 = nT-T, substituting in the above equations we get
𝑎𝑇 𝑎𝑇 𝑏𝑇
Y(nT) + y(nT) - [1 − ]y(nT-T) = [x(nT)+x(nT-T)]
2 2 2

z-transform of this difference equation is

𝑎𝑇 𝑎𝑇 𝑏𝑇
[1 + ]y(z) - [1 − ]z-1y(z) = [1+z-1]X(z)
2 2 2

System function of digital filter is

Finally

Comparing H(s) and H(z), the mapping from s-plane to the z-plane can be obtained as
2
s= ((1-z-1)/( 1+z-1)
𝑇

The relationship between s and z is known as bilinear transformation.

Let z = rejw and s = σ+jΩ
by solving and separating imaginary and real parts, we have

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

2 𝑤
by solving above equation Ω = 𝑇 tan 2
𝛺𝑇
w = 2tan-1 2

The warping effect:

Let Ω and w are frequency variables in the analog filter
2 𝑤
Ω = 𝑇 tan 2

For small values of w w = ΩT

The relationship Ω and w becomes non-linear and distortion is introduced in the frequency
scale of digital filter to that of analog filter. This is known as wrapping effect

Figure 5.7 Relationship between Ω

and w

Pre warping:
The warping effect can be eliminated by pre warping the analog filter. This can be done by
2 𝑤
Ω= tan 2
𝑇

Therefore we have

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Steps to design digital filter using bilinear transform technique:

1. From the given specifications, find pre warping analog frequencies using formula Ω =
2 𝑤
tan 2
𝑇

2. using the analog frequencies find H(s) of the analog filter.

3. Select the sampling rate of the digital filter, call it T seconds per sample.
4. Substitute value of s into the transfer function found in step-2.

Assignment cum Tutorial Questions

5.1. Butterworth Filter
Objective:
1. What is the lowest order of the Butterworth filter with a pass band gain 1 dB at ΩP=4
rad/sec and stop band attenuation greater than or equal to 20dB at ΩS = 8 rad/sec?
BL-V
a) 4 b) 5 c) 6 d) 3

2. What is the expression for cut-off frequency in terms of pass band gain? BL-III
𝜴𝒑 𝛺𝑝 𝛺𝑝 𝛺𝑝
a) b)(100.1 𝛼𝑝 +1)1/2𝑁 c)(100.1 𝛼𝑝 −1)1/𝑁 d) (100.1 𝛼𝑝 +1)1/𝑁
(𝟏𝟎𝟎.𝟏 𝜶𝒑 −𝟏)𝟏/𝟐𝑵

3. What is the equation for cut-off frequency in terms of stopband gain? BL-III
𝜴𝒔 𝑠𝛺 𝛺 𝛺
a)(𝟏𝟎𝟎.𝟏 𝜶𝒔 −𝟏) 𝟏/𝟐𝑵 b)(100.1 𝛼𝑠 +1)1/2𝑁
c)(100.1 𝛼𝑠𝑠−1)1/𝑁 d) (100.1 𝛼𝑠𝑠+1)1/𝑁

4. What is the order N of the low pass Butterworth filter? BL-II

100.1 𝛼𝑠 + 1 𝟏𝟎𝟎.𝟏 𝜶𝒔 − 𝟏
log √ 0.1 𝛼𝑝 100.1 𝛼𝑠 − 1 𝐥𝐨𝐠 √ 𝟎.𝟏 𝜶 100.1 𝛼𝑠 + 1
10 +1 log 0.1 𝛼𝑝 𝟏𝟎 𝒑 −𝟏 log 0.1 𝛼𝑝
10 −1 10 +1
a) 𝛺 b) 𝛺 c) 𝜴 d) 𝛺
log 𝑠 log 𝑠 𝐥𝐨𝐠 𝒔 log 𝑠
𝛺𝑝 𝛺𝑝 𝜴𝒑 𝛺𝑝

5. What is the cut-off frequency of the Butterworth filter with a pass band gain 1 dB at
ΩP=4 rad/sec and stop band attenuation greater than or equal to 20dB at ΩS=8 rad/sec?
BL-V
a) 3.5787 b) 1.069 c) 6 d) 4.5787
Descriptive:
1. Design an analog Butterworth filter that has -2dB pass band attenuation at a
frequency of 20 rad/sec and at least -10 dB stop band attenuation at 30 rad/sec.
BL-VI
𝜴𝒑 𝜴𝒔
2. Prove that 𝛺𝑐 = = (𝟏𝟎𝟎.𝟏 𝜶𝒔 −𝟏) BL-V
(𝟏𝟎𝟎.𝟏 𝜶𝒑 −𝟏)𝟏/𝟐𝑵 𝟏/𝟐𝑵

5.2. Chebyshev Filter (Type-I)

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Objective:
1. What is the value of Chebyshev polynomial of degree 1? BL-II
a) 1 b) x c) -1 d) –x
2. Which of the following defines a chebyshev polynomial of order N, CN(x)?
BL-I
a) cos(Ncos-1x) for all x b) cosh(Ncosh-1x) for all x
c) cos(Ncos-1x), |x|<1
cosh(Ncosh-1x), |x|>1 d) None of the mentioned

3. What is the equation for magnitude frequency response |H(jΩ)| of a low pass
Chebyshev-I filter? BL-I
1 𝟏
a) b)
𝛺 𝜴
2(
√1−𝜀 2 𝐶𝑁 𝛺
) √𝟏 + 𝜺𝟐 𝑪𝟐𝑵 (𝜴 )
𝑝 𝒑

1 𝟏
c) d)
𝛺 𝜴
2(
√1−𝜀 𝐶𝑁 𝛺
) √𝟏 + 𝜺 𝑪𝟐𝑵 (𝜴 )
𝑝 𝒑

4. The poles of H(s).H(-s) of Chebyshev filter are found to lie on ___ BL-IV
a) Circle b) Parabola c) Hyperbola d) Ellipse
Descriptive:
1. Given the specifications: pass band and stop attenuations are 3dB and 16dB at 1 KHz
and 2 KHz frequencies respectively. Determine the order of filter using Chebyshev
approximation. BL-V
2. Find the transfer function of Chebyshev filter with pass band and stop attenuations are
3dB and 16dB at 1 KHz and 2 KHz frequencies respectively.
BL-VI

5.3 Design of IIR filters from analog filters

Objective:
1. The following are the techniques for digitizing transfer function of analog filter
BL-I
a) Approximation of derivatives b) Impulse invariant transformation
c) Both a and b d) neither a nor b
2. By impulse invariance method, the IIR filter will have a unit sample response h(n) that is the
sampled version of the analog filter. BL-I
a) True b) False

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

3. When σ=0, then what is the condition on ‘r’? BL-II

a) 0<r<1 b) r=1 c) r>1 d) r=0
4. Which of the following filters cannot be designed using impulse invariance method?
BL-III
a) Low pass b) Band pass
c) Low and band pass d) High pass
5. In the Bilinear Transformation mapping, which of the following are correct? BL-II
a) All points in the LHP of s are mapped inside the unit circle in the z-plane
b) All points in the RHP of s are mapped outside the unit circle in the z-plane
c) All points of s are mapped inside & outside the unit circle in the z-plane
d) None of the mentioned
6. Is IIR Filter design by Bilinear Transformation is the advanced technique when compared to
other design techniques? BL-II
a) True b) False

Descriptive:
1. Design a third order Butterworth digital filter using Impulse Invariant Technique. Assume
Sample period T = 1sec and transfer function H(s) = 1/ [(s+1) (s2+s+1)]. BL-VI
2. Using bilinear transform, design a high pass filter monotonic in pass band with cut-off
Frequency of 1000Hz and down 10dB at 350 Hz. The sampling frequency is 5000Hz.
BL-VI

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

UNIT-V: Design of FIR Filters & Realization of Digital Filters

Objective: To introduce the design concepts of FIR Filters and the realization of IIR and FIR
filters

Syllabus: Design of FIR Filters: Linear phase FIR Filters- frequency Response, Fourier series
method of designing FIR Filters, Design of FIR Filters using windows (Rectangular, Bartlett,
Hanning and Hamming)

Realization of IIR filters-Direct form I, II. Realization of FIR filters - transversal structures,
cascade realization

Outcomes:
At the end of the Course, Student will be able to

➢ design FIR filters and realize IIR and FIR filters

6.1 Linear phase FIR filters – Frequency Response & The Fourier series
method of Designing FIR Filters
6.1.1 Linear phase structure for FIR filters

The transfer function of a FIR casual filter is given by

H(z) = ∑𝑁−1
𝑛=0 ℎ(𝑛)z
-1

Where h(n) = h(N-1-n)

Where h(n) is the impulse response of the filter

The Fourier transform of h(n) is

H(ejw) = ∑𝑁−1
𝑛=0 ℎ(𝑛)e
-jwn

For N = even
𝑁/2−1
H(z) = ∑𝑛=0 ℎ(𝑛)z-n+ ∑𝑁−1
𝑛=𝑛/2 ℎ(𝑛)z
-n

Substituting the h(n) value in above equation we get

(𝑁−2)/2
H(z) = ∑𝑛=0 ℎ(𝑛) [z-n+z-(N-1-n)]
6.1.2 Linear phase FIR filters – Frequency Response

An FIR filter has linear phase if its unit sample response satisfies the condition

h(n) = ± h ( N-1-n) n= 0, 1, 2…N-1

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Case I: Symmetrical impulse response, N odd

The frequency response of impulse response of H(z) can be written as

This can be split into

In general for N samples

Let n= N-1-m, we have

For a symmetrical impulse response h(n) = h(N-1-n), substitute it in the above

equation we get

Let (N-1)/2 – n =p, then

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Rewriting the above equation we can get

̅ ( ejw) is called as zero phase frequency response to distinguish it from the

𝐻
magnitude response.

The frequency response of symmetric impulse response for N odd and the
difference between H(ejw)and 𝐻
̅ (ejw) and between (w) and  H(ejw) are
shown in the figure

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Relation between magnitude response H(ejw)and the zero phase response 𝐻

̅
(ejw) and between (w) and  H(ejw)
Case-II: Symmetric impulse response for N even

H(ejw) = ∑𝑁−1
𝑛=0 ℎ(𝑛)e
-jwn

We know h(n) = h(N-1-n) then

(𝑁−2)/2 (𝑁−2)/2
H(ejw) = ∑𝑛=0 ℎ(𝑛)e-jwn + ∑𝑛=0 ℎ(𝑛)e-jw(N-1-n)
𝑁/2
H(ejw) = e-jw(N-1)/2∑𝑛=1 𝑏(𝑛)cos(n-1/2)w

Where b(n) = 2h[N/2-n]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The frequency response of linear phase filter with symmetric impulse response for N even is
shown in the figure 6.2

Figure 6.2 Frequency response of linear phase filter with symmetric impulse response for N
even

Case III: Anti-symmetric N odd

𝑁−1
For this type of sequence h[ ] =0
2

H(ejw) = ∑𝑁−1
𝑛=0 ℎ(𝑛)e
-jwn

(𝑁−3)/2 𝑁−1
H(ejw) = ∑𝑛=0 ℎ(𝑛)e-jwn +h[
2
] e-jw(N-1)/2 +∑(𝑁−1)
𝑛=(𝑁+1)/2 ℎ(𝑛)e
-jwn

We know that h(n) = h(N-1-n), therefore

(𝑁−3)/2 (𝑁−3)/2
H(ejw) = ∑𝑛=0 ℎ(𝑛)e-jwn -∑𝑛=(𝑁+1)/2 ℎ(𝑛)e-jw(N-1-n)

(𝑁−1)/2
= e-jw(N-1)/2 e-jπ/2 ∑𝑛=1 𝑐(𝑛)sinwn
𝑁−1
Where c(n) = 2 h[ -n], Therefore
2

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The frequency response of linear phase FIR filter for anti-symmetric sequence with N odd is
shown in the figure 6.3

Figure 6.3 Frequency response of linear phase FIR filter with antisymmetric sequence with N
odd

Case IV: N even

H(ejw) = ∑𝑁−1
𝑛=0 ℎ(𝑛)e
-jwn

(𝑁−2)/2 (𝑁−2)/2
H(ejw) = ∑𝑛=0 ℎ(𝑛)e-jwn +∑𝑛=0 ℎ(𝑁 − 1 − 𝑛)e-jw(N-1-n)

We know that h(n) = h(N-1-n), therefore

(𝑁−2)/2 (𝑁−2)/2
H(ejw) = ∑𝑛=0 ℎ(𝑛)e-jwn -∑𝑛=0 ℎ(𝑛)e-jw(N-1-n)

= e-jw(N-1)/2 e-jπ/2 [∑𝑁/2

𝑛=1 𝑑(𝑛)sinwn[n-1/2]]

𝑁
Where d(n) = 2 h[ -n]
2

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The frequency response of linear phase filter for antisymmetric impulse response with N even
is shown in figure 6.4

Figure 6.4 Frequency response of linear phase filter for antisymmetric impulse response with
N even

6.1.3 The Fourier series method of Designing FIR Filters

The frequency response H(ejw) of a system is periodic in 2π. From fourier series analysis we
know that any periodic function can be expressed as a linear combination of complex
exponentials. Therefore, the desired frequency response of an FIR filter can be expressed by
the Fourier series

Hd(ejw) = ∑∞
𝑛=∞ ℎd(n)e
-jwn

Where the Fourier coefficients hd(n) are the desired impulse response sequence of the filter

1 𝜋
Hd(n) = ∫ 𝐻d(ejw)ejwdw
2𝜋 −𝜋

The z-transform of the sequence is given by

H(z) = ∑∞
𝑛=∞ ℎ d(n)z
-n

To get an FIR filter transfer function, the series can be truncated by assigning

H(n) = hd(n) for |n| ≤ (N-1)/2

= 0 otherwise
(N−1)/2
Then H(z) = ∑ℎ=−[(𝑁−1)/2] ℎd(n)z-n

(N−1)/2
= h(0) + ∑𝑛=1 [ℎ(n)z-n+h(-n)zn]
For a symmetrical impulse response having symmetry at n=0 then h(-n) = h(n), Therefore

(N−1)/2
H(z) = h(0) + ∑𝑛=1 [ℎ(n)[z-n+zn]
Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The above transfer function is not physically realizable. Realizability can be brought by
multiplying the above eq by z-(N-1)/2 when (N-1)/2 is delay in samples

H’(z) = z-(N-1)/2 H(z)

(N−1)/2
= z-(N-1)/2[h(0) + ∑𝑛=1 [ℎ(n)(z-n+zn)]]

Assignment cum Tutorial Questions

6.1 Linear phase FIR filters – Frequency Response & The Fourier series method of
Designing FIR Filters:

Objective:

1. Ideal filters are BL-II

a) Causal b) Non-Causal c) may be causal or non-causal d) none of these
2. The frequency response of digital filters is BL-I
a) Periodic b) Non-periodic c) may be periodic or non-periodic d) none of these
3. Which of the following condition should the unit sample response of a FIR filter satisfy
to have a linear phase? BL-II
a) h (N – 1 - n) n = 0, 1, 2, …, N-1 b) ± h (N – 1 - n) n = 0, 1, 2, …, N-1
c) – h (N – 1 - n) n = 0, 1, 2, …, N-1 d) None of the mentioned
4. What is the value of h (N – 1 / 2) if the unit sample response is anti-symmetric? BL-I
a) 0 b) 1 c) -1 d) None of the mentioned
5. The anti-symmetric condition with M even is not used in the design of which of the
following linear-phase FIR filter? BL-II
a) Low pass b) High pass c) Band pass d) Band stop
6. What is the equation for phase delay of an FIR filter? BL-II
a) −𝜽(𝝎)/𝝎 b) −𝑑𝜃(𝜔)/𝑑𝜔 c) 𝜃(𝜔)/𝜔 d) 𝑑𝜃(𝜔)/𝑑𝜔
7. Determine Frequency response of FIR filter defined by y (n) = 0.25x(n) + x(n-1) +
0.25x(n-2) BL-V
a) H (w) = 0.52 - e-j2w + 0.52e5jw b) H (w) = 0.25+e-6jw - 0.25ejw
c) H (w) = 0.25+e-jw + 0.25e-2jw d) H (w) = 0.52-e2jw - 0.52ejw

Descriptive:

1. Explain frequency response of FIR filter for the case of symmetrical impulse response,
and
N is odd. BL-IV
2. Design an ideal low pass filter with a frequency response

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

𝜋 𝜋
= 1 𝑓𝑜𝑟 − ≤𝜔≤
𝐻𝑑 (𝑒 𝑗𝜔 ) 2 2
𝜋
= 0 𝑓𝑜𝑟 ≤𝜔≤𝜋
2
Find the values of h(n) for N=11. Find H(z) and plot the magnitude response BL-VI

3. Explain Fourier series method of designing FIR filters. BL-IV

4. Distinguish between FIR and IIR filters. BL-II

5. What are the conditions for the impulse response of FIR filter to satisfy for

a) constant group and phase delay b) only constant group delay BL-IV

6. Determine the frequency response of FIR filter defined by y(n) = 0.25 x(n) + x(n-1)

+ 0.25 x(n-2). Calculate the phase delay and group delay. BL-V

6.2 Design of FIR filters using windows

Window having very small main lobe width with most of the energy contained with it (i.e.,
ideal window frequency response must be impulsive).Window design is a mathematical
problem, more complex the window lesser are the distortions. Rectangular window is one of
the simplest windows in terms of computational complexity. Windows better than rectangular
window are, Hamming, Hanning, Blackman, Bartlett, Triangular and Kaiser. The different
window functions are discussed in the following section.

The desired frequency response Hd(ejw) of a filter can be expanded in terms of Fourier series.

Where

𝑁−1
One way of obtaining FIR filter is to truncate the infinite Fourier series at n = ± [ ], where
2

N is length of the sequence. But the abrupt truncation of the Fourier series results in oscillation
in the pass band and stop
band due to slow coverage
of Fourier series, to reduce

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

these oscillations Fourier coefficients of the filter are modified by multiplying the infinite
impulse response with a finite weighing sequence w(n) called a window where

h(n) is the finite duration sequence when multiplying w(n) and hd(n)

The frequency response H(ejw) of the filter is obtained by convolution of Hd(ejw) and W(ejw)
is given by

Figure 6.5 different window techniques

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The window truncation the infinite impulse response should have some desirable
characteristics they are

1. The central lobe of the frequency response of the window should contain most of the
energy and should be narrow.

2. The highest side lobe level of the frequency response should be small.

3. The side lobes of the frequency response should decrease in energy rapidly as w tends to π.

Rectangular window

The rectangular window sequence is given by

wR(n) = 1 for –(N-1)/2 ≤ n ≤ (N-1)/2

= 0 otherwise

Example is shown in the figure 6.6 for N = 25

Figure 6.6 Rectangular window

The spectrum of rectangular window is given by

Finally we get
𝑤𝑁
𝑠𝑖𝑛
2
=
sin 𝑤/2

The spectrum sequence for N =25 is shown in the figure6.7

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure 6.7 (a) frequency response of rectangular window for N = 25 (b) log magnitude
response of rectangular window for N = 25

The frequency response of the truncated filter can be obtained by periodic convolution ig
given by

Triangular or Bartlett window

The N-point triangular window is given by

The Fourier transform of the triangular window is

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The frequency spectrum for N = 25 are shown in the figure 6.8

Figure 6.8 Triangular window sequence

Figure (a) Frequency response of triangular window for N= 25 (b) log magnitude response of
triangular window for N = 25.

Hanning window

The hanning window can be obtained from the Raised cosine window by substituting α = 0.5

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The Frequency response is given by

Figure 6.10 (a) Frequency response of Hanning window for N = 25 (b) log magnitude
response of Hanning window for N = 25

Hamming window

The hamming window can be obtained from the Raised cosine window by substituting α =
0.54

wH(n) = 0.54 + 0.46 cos(2πn/N-1) for –(N-1)/2 ≤ n ≤ (N-1)/2

= 0 Otherwise

The frequency response of Hamming window is given by

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Figure 6.12 (a) Frequency response of Hamming window for N= 25 (b) log magnitude
response of Hamming for N = 25

Assignment cum Tutorial Questions

6.2 Design of FIR filters using windows:

Objective:

1. What is the width of the main lobe of the frequency response of a rectangular window
of length M-1? BL-II
a) π/N b) 2π/N c) 4π/N d) 8π/N
2. With an increase in the value of N, the height of each side lobe ____________ BL-II
Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25
R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

a) Do not vary b) Does not depend on value of N

c) Decreases d) Increases
2|𝑛|
3. Which of the following windows has a time domain sequence ℎ(𝑛) = 1 − 𝑁−1 ?BL-II

a) Bartlett window b) Blackman window

c) Hanning window d) Hamming window
4. What is the approximate transition width of main lobe of a Hamming window? BL-II
a) 4π/N b) 8π/N c) 12π/N d) 2π/N
5. What is the peak side lobe (in dB) for a rectangular window? BL-II
a) -13 b) -27 c) -32 d) -58
6. The oscillatory behavior near the band edge of the low pass filter is known as Gibbs
phenomenon BL-II
a) True b) False
7. How does the frequency of oscillations in the pass band of a low pass filter varies with
the value of N? BL-II
a) Decrease with increase in N b) Increase with increase in N
c) Remains constant with increase in N d) None of the mentioned

Descriptive:

1. Discuss the procedure for designing FIR filters using rectangular and Bartlett windows
BL-IV

2. Design an ideal high pass filter with a frequency response

𝜋
= 1 𝑓𝑜𝑟 ≤𝜔≤𝜋
𝐻𝑑 (𝑒 𝑗𝜔 ) 4
𝜋
= 0 𝑓𝑜𝑟 𝜔 ≤
4
using Hanning window for N = 11. BL-VI

3. Design an ideal high pass filter with a frequency response

𝜋
= 1 𝑓𝑜𝑟 ≤𝜔≤𝜋
𝐻𝑑 (𝑒 𝑗𝜔 ) 4
𝜋
= 0 𝑓𝑜𝑟 𝜔 ≤
4
using Hamming window for N = 11. BL-VI

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Assignment cum Tutorial Questions

6.3 Realization of Digital Filters – IIR Filters

o Direction-Form I Realization

The digital filter transfer function is given by

𝑌(𝑧) 𝑏0 + 𝑏1 𝑧 −1 + ⋯ + 𝑏𝑀 𝑧 −𝑀 𝐵(𝑧)
𝐻(𝑧) = = =
𝑋(𝑧) 1 + 𝑎1 𝑧 −1 + ⋯ + 𝑎𝑁 𝑧 −𝑁 𝐴(𝑧)

This expression can also be written as

𝑌(𝑧) = 𝐻(𝑧)𝑋(𝑧)

Or

𝑏0 + 𝑏1 𝑧 −1 + ⋯ + 𝑏𝑀 𝑧 −𝑀
𝑌(𝑧) = ( ) 𝑋(𝑧)
1 + 𝑎1 𝑧 −1 + ⋯ + 𝑎𝑁 𝑧 −𝑁

Taking the inverse z-transform of the above equation yields the difference equation

𝑦(𝑛) = 𝑏0 𝑥(𝑛) + 𝑏1 𝑥(𝑛 − 1) + ⋯ + 𝑏𝑀 𝑥(𝑛 − 𝑀)

−𝑎1 𝑦(𝑛 − 1) − 𝑎2 𝑦(𝑛 − 2) − ⋯ − 𝑎𝑁 𝑦(𝑛 − 𝑁)

This difference equation can be implemented by a direct-form I realization. We introduce the

following notation:

𝒃𝒊
𝒙(𝒏) 𝒃𝒊 ∙ 𝒙(𝒏) 𝒙(𝒏) 𝒙(𝒏 − 𝟏)
𝒛−𝟏

The direct form I realization of the above difference equation is given in the figure below.

The direct form I realization of the second order IIR filter (𝑁 = 𝑀 = 2) is given by

𝑦(𝑛) = 𝑏0 𝑥(𝑛) + 𝑏1 𝑥(𝑛 − 1) + 𝑏2 𝑥(𝑛 − 2) − 𝑎1 𝑦(𝑛 − 1) − 𝑎2 𝑦(𝑛 − 2)

is shown in the diagram below:

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Direction-Form II Realization

Using the digital filter transfer function, we can write

𝐵(𝑧) 𝑋(𝑧)
𝑌(𝑧) = 𝐻(𝑧)𝑋(𝑧) = 𝑋(𝑧) = 𝐵(𝑧) ( )
𝐴(𝑧) 𝐴(𝑧)

This expression can also be written as

𝑋(𝑧)
𝑌(𝑧) = (𝑏0 + 𝑏1 𝑧 −1 + ⋯ + 𝑏𝑀 𝑧 −𝑀 ) ( ) (1)
1 + 𝑎1 𝑧 −1 + ⋯ + 𝑎𝑁 𝑧 −𝑁

By defining

1
𝑊(𝑧) = 𝑋(𝑧) (2)
1 + 𝑎1 𝑧 −1 + ⋯ + 𝑎𝑁 𝑧 −𝑁

We have

𝑌(𝑧) = (𝑏0 + 𝑏1 𝑧 −1 + ⋯ + 𝑏𝑀 𝑧 −𝑀 )𝑊(𝑧) (3)

The corresponding difference equations for equations (2) and (3) are

𝑤(𝑛) = 𝑥(𝑛) − 𝑎1 𝑤(𝑛 − 1) − 𝑎2 𝑤(𝑛 − 2) − ⋯ − 𝑎𝑁 𝑤(𝑛 − 𝑁) (4)

and

𝑦(𝑛) = 𝑏0 𝑤(𝑛) + 𝑏1 𝑤(𝑛 − 1) + ⋯ + 𝑏𝑀 𝑤(𝑛 − 𝑀) (5)

These difference equations can be implemented by a direct-form II realization.

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

The direct form I realization of the second order IIR filter (𝑁 = 𝑀 = 2) is given by

𝑦(𝑛) = 𝑏0 𝑥(𝑛) + 𝑏1 𝑥(𝑛 − 1) + 𝑏2 𝑥(𝑛 − 2) − 𝑎1 𝑦(𝑛 − 1) − 𝑎2 𝑦(𝑛 − 2)

is shown in the diagram below:

6.3 Realization of Digital Filters – IIR Filters:

Objective:

1. If M and N are the orders of numerator and denominator of rational system function
respectively, then how many multiplications are required in direct form-I realization of
that IIR filter?
BL-II
a) M+N-1 b) M+N c) M+N+1 d) M+N+2
2. If M and N are the orders of numerator and denominator of rational system function
respectively, then how many memory locations are required in direct form-II realization
of that IIR filter?
BL-II
a) M+N+1 b) M+N c) Min [M,N] d) Max [M,N]

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

3. If M and N are the orders of numerator and denominator of rational system function
respectively, then how many memory locations are required in direct form-I realization
of that IIR filter? BL-II
a) M+N+1 b) M+N c) M+N-1 d) M+N-2

Descriptive:

1. Explain the following methods of realization of IIR filters

a) Direct form I b) Direct form II BL-II

2. Obtain the direct form-I realization for the system described by difference equation

y(n) = 0.5y(n-1)-0.25y(n-2)+x(n)+0.4x(n-1) BL-V

4. Determine the direct form-II for realization of the system

y(n) = -0.1y(n-1)+0.72y(n-2)+0.7x(n)-0.252x(n-2) BL-V

6.4 Realization of digital filters – FIR filters

Transversal Structure

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

Cascade (Series) Realization

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25

R -20 DIGITAL SIGNAL PROCESSING III B. Tech-I Sem

6.4 Realization of digital filters – FIR filters:

Objective:

1. In general, an FIR system is described by the difference equation BL-II

𝑁−1

𝑦(𝑛) = ∑ 𝑏𝑘 𝑥(𝑛 − 𝑘)
𝑘=0

a) True b) False
2. What is the general system function of an FIR system? BL-II
a) ∑𝑁−1
𝑘=0 𝑏𝑘 𝑥(𝑛 − 𝑘) b) ∑𝑵−𝟏
𝒌=𝟎 𝒃𝒌 𝒛
−𝒌

c) ∑𝑁
𝑘=0 𝑏𝑘 𝑧
−𝑘
d) None of the mentioned

3. Which of the following filters have a cascade realization as shown below? BL-III

a) IIR filter b) Comb filter c) High pass filter d) FIR filter

Descriptive:

1. Discuss the following methods of realization of FIR filters

a) Transversal structure b) Cascade realization BL-II

2. Determine the direct form realization of system function H(z) = 1 + 2z-1 - 3z-2 - 4z-3 +
5z-4 BL-V

3. Obtain the cascade realization of system function H(z) =(1 + 2z-1 - z-2) ( 1 + z-1 - z-2)

BL-V

Seshadri Rao Gudlavalleru Engineering College Dept. of ECE AS&A A. Y: 2024-25