Scribd
Upload
96 views2 pages

EC60085-DSP and Its Applications Homework-2: Due Date: 14 Sept. 2019

1. This document contains 9 questions regarding digital signal processing (DSP) and its applications. The questions cover topics such as determining output noise variance of IIR filters, realizing filters in parallel and cascade forms, calculating passband and stopband ripple values, deriving digital filters from analog prototypes using impulse invariance, and performing spectral transformations. 2. Question 1 asks to determine the output noise variance of an IIR filter due to input quantization noise. Question 2 gives a parallel form filter and asks for the output noise variance due to input quantization. 3. Questions 3-6 cover realizing filters in cascade and parallel forms, calculating filter design parameters like passband ripple and stopband attenuation, and analyzing

Uploaded by

harshit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
96 views2 pages

EC60085-DSP and Its Applications Homework-2: Due Date: 14 Sept. 2019

1. This document contains 9 questions regarding digital signal processing (DSP) and its applications. The questions cover topics such as determining output noise variance of IIR filters, realizing filters in parallel and cascade forms, calculating passband and stopband ripple values, deriving digital filters from analog prototypes using impulse invariance, and performing spectral transformations. 2. Question 1 asks to determine the output noise variance of an IIR filter due to input quantization noise. Question 2 gives a parallel form filter and asks for the output noise variance due to input quantization. 3. Questions 3-6 cover realizing filters in cascade and parallel forms, calculating filter design parameters like passband ripple and stopband attenuation, and analyzing

Uploaded by

harshit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

EC60085-DSP and Its Applications

Homework-2
Due Date: 14th Sept. 2019
1. Determine the output noise variance due to the propagation of the input quantization noise for the
following IIR digital filter:
(4𝑧 + 1)(𝑧 2 − 4𝑧 + 1)
𝐻(𝑧) =
(𝑧 − 0.2)(𝑧 + 0.4)(𝑧 2 + 0.75𝑧 + 0.4)
2. Determine the expression for the normalized output noise variance due to input quantization of
the following digital filter in a parallel form
𝐵 𝐶
𝐻(𝑧) = 𝐴 + +
𝑧+𝛽 𝑧+𝛾
Each section in the parallel structure is realized in DF-II form. What is the value of the variance
for 𝛽 = 0.2, 𝛾 = −0.75, 𝐴 = 6, 𝐵 = −2, 𝐶 = 3?

3. Realize 𝐻(𝑧) in Problem 2 above as SoS and DF-II implementation. Scale the structure using the
L2-norm and then compute the output noise variance due to product round-off, assuming
a. quantization of products before addition
b. quantization after addition
4. Determine the peak ripple values 𝛿𝑝 and 𝛿𝑠 for each of the following sets of peak passband ripple
𝛼𝑝 and minimum stopband attenuation 𝛼𝑠 :
i. 𝛼𝑝 = 0.24 dB, 𝛼𝑠 = 49 dB ii. 𝛼𝑝 = 0.14 dB, 𝛼𝑠 = 68 dB
5. Determine the peak passband ripple 𝛼𝑝 and minimum stopband attenuation 𝛼𝑠 in dB for each of
the following sets of peak ripple values 𝛿𝑝 and 𝛿𝑠 :
i. 𝛿𝑝 = 0.04, 𝛿𝑠 = 0.08 ii. 𝛿𝑝 = 0.015, 𝛿𝑠 = 0.04
6. Let H(z) be the transfer function of a lowpass digital filter with passband edge at 𝜔𝑝 , stopband
edge at 𝜔𝑠 , passband ripple of 𝛿𝑝 , and stopband ripple of 𝛿𝑠 . Consider a cascade of two identical
filters with a transfer function H(z). What are the passband and stopband ripples of the cascade
at 𝜔𝑝 and 𝜔𝑠 , respectively? Generalize the results for a cascade of M identical sections.
7. The impulse invariance method is another approach to the design of a causal IIR digital filter
𝐺(𝑧) based on the transformation of a prototype causal analog transfer function 𝐻𝑎 (𝑠). If ℎ𝑎 (𝑡)
is the impulse response of 𝐻𝑎 (𝑠), in the impulse invariance method, we require that the unit
sample response 𝑔[𝑛] of 𝐺(𝑧) be given by the sampled version of ℎ𝑎 (𝑡) sampled at uniform
intervals of 𝑇 seconds, i.e.,
ℎ𝑎 (0+)
, 𝑛 = 0,
𝑔[𝑛] = { 2
ℎ𝑎 (𝑛𝑇), 𝑛 ≥ 1.
(a) Show that 𝐺(𝑧) and 𝐻𝑎 (𝑠) are related through
∞
𝑍−𝑡𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚{𝑔[𝑛]} 1 2𝜋𝑘
𝐺(𝑧) ⇒ ∑ 𝐻𝑎 (𝑠 + 𝑗 )|
𝑇 𝑇 𝑠=(1) ln 𝑧
𝑘=−∞ 𝑇
1
(b) Show that the transformation 𝑠 = 𝑇 ln 𝑧 has all the desirable properties of mapping from 𝑠-
domain to 𝑧-domain.
(c) Develop the convolution under which the frequency response 𝐺(𝑒 𝑗𝜔 ) of 𝐺(𝑧) will be a scaled
replica of the frequency response 𝐻𝑎 (𝑗Ω) of 𝐻𝑎 (𝑠).
EC60085-DSP and Its Applications

(d) Show that the normalized digital angular frequency  is related to the analog angular
frequency  as 𝜔 = Ω𝑇.
8. Determine the digital transfer function obtained by transforming the following causal analog
transfer function using the impulse invariance method. Assume 𝑇 = 0.25 sec.
2(𝑠 + 2)
𝐻𝑎 (𝑠) =
(𝑠 + 3)(𝑠 2 + 4𝑠 + 5)
9. A second order lowpass IIR digital filter with a 3-dB cutoff frequency at 𝜔𝑐 = 0.55𝜋 has a
transfer function
0.3404(1 + 𝑧 −1 )2
𝐺𝐿𝑃 (𝑧) =
1 + 0.1842𝑧 −1 + 0.177𝑧 −2
Design a second-order lowpass filter 𝐻𝐿𝑃 (𝑧) with a 3-dB cutoff frequency at 𝜔̂𝑐 = 0.27𝜋 by
transforming the above lowpass transfer function using a lowpass-to-lowpass spectral
transformation.

You might also like