Broad Question:

Q) Write down the differences between FIR & IIR filter.

Ans:
FIR IIR
1) FIR filter stands for finite 1) IIR filter stands for infinite
impulse response filter. impulse response filter.
2) An FIR filter provides impulse 2) An IIR filter gives impulse
responses for a finite duration of responses for an infinite duration
time. of time.
3) Feedback system is not present 3) Feedback system is available in
in FIR filter. IIR filter.
4) The FIR filter is non-recursive 4) The IIR filter is recursive.
in nature.
5) FIR filters do not contain any 5) IIR filters are less stable than
poles. Hence the FIR filter is that of FIR filter contains both
always stable and the stability is poles and zero.
more than that of IIR filter.
6) The transfer function of FIR 6) The transfer function of IIR
filter contains only zeros. filter contains both poles and
zeros.
7) Due to the absence of feedback 7) The memory requirements of an
path in the FIR filters, it needs IIR filter is less than that of FIR
more memory to keep the more filter as it contains feedback path.
number of coefficients even for a
single function.
8) Differentiator, Hilbert 8) Butterworth filter, Chebyshev
transformer etc. type filter etc.
9) The uses of FIR filters are 9) The uses of IIR filter are found
found in audio system, control in biomedical sensor,
system, medical devices etc. telecommunication, audio
equalization etc.
Question
What is the frequency response of the following network?
[Supplementary Problem: 8.43]

Answers:
According to given diagram-
𝑌(𝑧) = 𝑋(𝑧)[0.5(𝑧 + 𝑧 ) + 0.2(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 )]
( )
𝑜𝑟, = 0.5(𝑧 + 𝑧 ) + 0.2(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 )
( )

𝑜𝑟, 𝐻(𝑧) = 0.5(𝑧 + 𝑧 ) + 0.2(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 ) + 0.1(𝑧 +𝑧 )

𝑜𝑟, ℎ(𝑛) = 0.5[𝛿 (𝑛 − 0) + 𝛿 (𝑛 − 7)] + 0.2[𝛿 (𝑛 − 1) + 𝛿 (𝑛 − 6)]
+0.1[𝛿(𝑛 − 2) + 𝛿(𝑛 − 5)] + 0.1[𝛿(𝑛 − 3) + 𝛿(𝑛 − 4)]
[Taking Inverse Z-Transform both side]

𝑜𝑟, ℎ 𝑒 = 0.5 𝑒 +𝑒 + 0.2 𝑒 +𝑒 + 0.1 𝑒 +𝑒 + 0.1[𝑒 +𝑒 ]

[Taking DTFT both side]

. . . . . . .
𝑜𝑟, ℎ 𝑒 =𝑒 0.5 𝑒 +𝑒 + 0.2 𝑒 +𝑒 + 0.1 𝑒 +𝑒
. .
+ 0.1 𝑒 +𝑒

. . . .
.
𝑒 +𝑒 𝑒 +𝑒
𝑜𝑟, ℎ 𝑒 =𝑒 1 + 0.4
2 2
.
𝑒 +𝑒 . 𝑒 .
+𝑒 .
+ 0.2 + 0.2
2 2

. [cos(3.5𝜔) + 0.4 cos(2.5𝜔) + 0.2 cos(1.5𝜔) + 0.2 cos(0.5𝜔)]

𝑜𝑟, ℎ 𝑒 =𝑒

∴ 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑔𝑖𝑣𝑒𝑛 𝑛𝑒𝑡𝑤𝑜𝑟𝑘,

 .
h e =e [cos (3.5ω) +0.4 cos (2.5ω) +0.2 cos (1.5ω) +0.2 cos (0.5ω)] (Ans)
Consider the causal linear shift-invariant filter with system function

Draw a signal flowgraph for this system using

(a) Direct form |
(b) Direct form ||

Answer:
a) it follows that the direct form I realization of H(z)is as follows:
b) For direct realization of H (z),we have
 Questions
1. Consider the causal linear shift-invariant filter with system function
1 + 0.875𝑧 −1
𝐻 (𝑧) =
(1 + 0.2𝑧 −1 + 0.9𝑧 −2 )(1 − 0.7𝑧 −1 )

Draw a signal flow graph for this system using

a. Direct form I
b. Direct form II
c. A cascade of first- and second-order systems realized in direct
form II
d. A cascade of first- and second-order systems realized in
transposed direct form II

Answer
(a)
1+0.875𝑧 −1
Given system function, 𝐻 (𝑧) =
(1+0.2𝑧 −1 +0.9𝑧 −2 )(1−0.7𝑧 −1 )

Direct Form I realization of H(z) will be-

 (b)
1+0.875𝑧 −1
Given system function, 𝐻 (𝑧) =
(1+0.2𝑧 −1 +0.9𝑧 −2 )(1−0.7𝑧 −1 )

Direct Form II realization of H(z) will be-

(c)
1+0.875𝑧 −1
Given system function, 𝐻 (𝑧) =
(1+0.2𝑧 −1 +0.9𝑧 −2 )(1−0.7𝑧 −1 )

1 1+0.875𝑧 −1
𝐿𝑒𝑡, 𝐻1 (𝑧) = & 𝐻2 (𝑧) =
(1−0.7𝑧 −1 ) (1+0.2𝑧 −1 +0.9𝑧 −2 )

∴ 𝐻 (𝑧) = 𝐻1 (𝑧)𝐻2 (𝑧)

Then Cascade Form realization of H(z) will be-

 (d)
If we change the direct form II systems in part (c) to transposed direct
form 11, we have the realization shown below.
Question:
Find the unit sample response for the following network:

Answer:
According to given diagram-

𝑌(𝑧) = 𝑋(𝑧)[0.2𝑧 + 0.4𝑧 + 0.1𝑧 + 0.5𝑧 + 0.3𝑧 ]

( )
𝑜𝑟, = 0.2𝑧 + 0.4𝑧 + 0.1𝑧 + 0.5𝑧 + 0.3𝑧
( )

𝑜𝑟, 𝐻(𝑧) = 0.2𝑧 + 0.4𝑧 + 0.1𝑧 + 0.5𝑧 + 0.3𝑧

𝑜𝑟, ℎ(𝑛) = 0.2𝛿(𝑛 − 0) + 0.4𝛿(𝑛 − 1) + 0.1𝛿(𝑛 − 2) + 0.5𝛿(𝑛 − 3) + 0.3𝛿(𝑛 − 4)
[Taking Inverse Z-Transform both side]

∴ 𝑈𝑛𝑖𝑡 𝑠𝑎𝑚𝑝𝑙𝑒 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑔𝑖𝑣𝑒𝑛 𝑛𝑒𝑡𝑤𝑜𝑟𝑘,

ℎ(𝑛) = 0.2𝛿(𝑛) + 0.4𝛿(𝑛 − 1) + 0.1𝛿(𝑛 − 2) + 0.5𝛿(𝑛 − 3) + 0.3𝛿(𝑛 − 4) (Ans)
 Student ID: 180908

Digital and Signal Processing Question & Answer

Question: How the cascade structure can be derived? Implement a sixth-order IIR filter as a
cascade of three second order system in direct form II with necessary figure.
Answer:
The cascade structure can be derived by factoring the numerator and denominator polynomials of
H(z)

This factorization corresponds to a cascade form of first order filters, each having one pole and
one zero. In general, the coefficient αk and βk will be complex. However, if h(n) is real, the roots
of H(z) will occur in complex conjugate pairs, and these complex conjugate factors may be
combined to form second order factors with real coefficient:

A sixth-order IIR filter implemented as a cascade of three second-order systems in direct form II
is shown in the fig 1:

Fig: A sixth-order IIR filter implemented as a cascade of three second-order systems

There is considerable flexibility in how a system may be implemented in cascade form. For
example, there are different pairings of the poles and zeros and different ways in which the
sections may be ordered.
 Q) For the network given below,
a. Find the system function and the unit sample response
b. Draw an equivalent direct form II structure

Answer
(a)
According to figure given in question,
1+2𝑧 −1 1
The system function 𝐻 (𝑧) = 2 + 1 + 1
1+2𝑧 −1 1−2𝑧 −1
(Ans)
Taking Z-transform both side, we get-
1 𝑛
The unit sample response, ℎ(𝑛) = 2𝛿 (𝑛) + (− ) 𝑢(𝑛) +
2
1 𝑛−1 1 𝑛
2 (− ) 𝑢(𝑛 − 1) + ( ) 𝑢(𝑛)
2 2

(Ans)
 (b)
We have got,
1+2𝑧 −1 1
𝐻 (𝑧) = 2 + 1 + 1
1+2𝑧 −1 1−2𝑧 −1

1 1 1 1 3
2(1+2𝑧 −1 )(1+2𝑧 −1 )+(1+2𝑧 −1 )(1−2𝑧 −1 )+1+2𝑧 −1 4+𝑧𝑧 −1 −2𝑧 −2
𝑜𝑟, 𝐻 (𝑧) = 1 1 = 1
(1+ 𝑧 −1 )(1+ 𝑧 −1 ) 1− 𝑧 −2
2 2 4

Then, the direct Form II realization of H(z) will be-

uwtion 50

valu ot u ntt Samp

un
t, 4 ist 4hu
cawa iHert be
wsponse 6 a disiru

hg(0)=3 ha(1) = and h(2) =

es-2sreiznt
mathod,find +h
wir Ina Pada
opproximation

pole t e r t
that ha, a
52Con& ortdert all
o a

5ueh 4hat
unt 5ampl r p o n s h (n),
1, 2
h (m)= ha (n) Jortn 0,

4 h 2uation
tort a second oredart all pol i+ert

apprmoxi mation r
or th Pao

1
hy Co) O

ha1) hd Co) a(1) 0

( ) h a ( 1) ho)
 hih volw', tor h (n)
hy (7)
h he iven

becom

a(1)

tmom h last 2 2quation wa hav

a(1)

6oing tort al1) and a(1) we ind

a )- aD= -

hn ing t rdt 2quation w have

blo) =
Th yhem unhion Jor 4 IHe i
Q)What do you mean by wiener filter? Write some application of wiener filter.

In signal processing, the Wiener filter is a filter used to produce an estimate of a

desired or target random process by linear time-invariant (LTI) filtering of an
observed noisy process, assuming known stationary signal and noise spectra, and
additive noise. The Wiener filter minimizes the mean square error between the
estimated random process and the desired process.

Wiener filters play a central role in a wide range of applications such as-
1. linear prediction,
2. echo cancellation,
3. signal restoration,
4. channel equalization and
5. system identification.
The Wiener filter coefficients are calculated to minimize the average squared
distance between the filter output and a desired signal.