National Institute of Technology Rourkela

NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA
Department of Electrical Engineering

End-Semester Examination, 2019 - 2020 (Autumn)
Program: B.Tech
Semester: 5th
Subject Name: Digital Signal Processing
Subject Code: EE3403
Number of Pages: 2 Full Marks: 50 Duration: 3 Hours
Answer all questions. Figures at the right hand margin indicate marks.
Answer must be brief and to the point with clear mention of necessary assumption.
1. (a) Let x[n] be the sequence with pole-zero plot as shown in the Figure below.
1 n
Sketch the pole-zero plot of x[n] (3)
2
(b) A discrete-time LTI system is described by the difference equation:

3
y[n] = y[n − 1] + y[n − 2] + x[n − 1]
2
Determine a stable (non-causal) impulse response that satisfies the given difference equation. (4)
(c) Consider the spectrum of an analog signal as shown below. Determine the minimum sampling
frequency required to avoid aliasing if fL = 90 M Hz and fH = 110 M Hz. (3)
2. (a) Compare the computational complexity of 1024-point FFT (radix-2 Decimation-in-Frequency

algorithm) with 1024-point DFT (direct computation)? Clearly mention the number of addition
and complex multiplication required for each case. (3)
(b) We use the DFT to compute the magnitude spectrum of a sampled data sequence with a
sampling rate fs = 20 kHz. The desired frequency resolution is 0.5 Hz. Determine the number
of data points would be used by the radix-2 FFT algorithm and actual frequency resolution in
Hz, assuming that the sufficient number of data samples are available for processing. (3)
Question 2 continues on the next page. . .

Digital Signal Processing End-Semester Examination Page 2 of 2
(c) Consider a causal discrete-time LTI system whose output y[n] and input x[n] are related by a
difference equation as follows:
1 1
y[n] − y[n − 2] = x[n − 2] − x[n]
4 4
Let x[n] and y[n] are two causal N-point
P −1 sequences. Their values are zero outside
PN the range
−1
0 ≤ n ≤ N − 1. It is also given that N n=0 |x[n]| 2 =10, then determine the value of
n=0 |y[n]|2. (4)
3. (a) Find the 3-dB bandwidth of the prototype low-pass filter with system function: (3)
Hp (z) = 0.5(1 + z −1 )
(b) The pole-zero plots of two different causal LTI systems are shown in the following figure. Which
of the following properties apply to each of the systems: Stable, IIR, FIR, Minimum-phase and
All-pass. (4)
(c) Suppose an audio processing system requires a sampling rate conversion scheme with the
following specifications:
Input sampling rate = 16 kHz and Output sampling rate = 44 kHz
Design this sampling rate conversion system by cascading an interpolator and a decimator
(Block-diagram only). Specify the cut-off frequencies and gains of filters which are to be used. (3)
4. (a) A linear-phase FIR system has a real impulse response h[n] whose Z-transform is known to have
the form:
H(z) = (1 − az −1 )(1 − ejπ/2 z −1 )(1 − bz −1 )(1 − 0.5z −1 )(1 − cz −1 )
where a,b and c are zeros of H(z) that you are to find. It is also known that H(ejω )=0 for ω=0.
(i) Determine the impulse response. (ii) Compute the group-delay of the system. (4)
(b) Consider a IIR filter with transfer function as follows:
1 + 2z −1
H(z) =
1 − 1.5z −1 + 0.9z −2
Write the difference equation of this system and also draw the Direct Form-II structure. (3)
(c) Given an FIR filter transfer function: H(z) = 0.2 + 0.5z −1 − 0.3z −2 + 0.5z −3 + 0.2z −4
perform the linear-phase FIR filter realization with minimal resource utilization. (3)
5. (a) Write down the basic advantages of FIR filters over IIR filters (Two valid points are sufficient). (2)
(b) What is ‘Gibbs Phenomenon’ ? Explain why does it occur during FIR filter design using
rectangular window? (4)
(c) Consider the Butterworth low-pass prototype filter with the transfer function:
1
Ha (s) =
s+1
Determine the system function H(z) of the discrete-time filter designed from Ha (s) based on
the step response invariance transformation (Assume sampling frequency = 10 Hz). (4)
End of question paper. ALL THE BEST.

National Institute of Technology Rourkela

Uploaded by

Copyright:

Available Formats

National Institute of Technology Rourkela

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

National Institute of Technology Rourkela

Uploaded by

Copyright:

Available Formats

NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA

Department of Electrical Engineering

Number of Pages: 2 Full Marks: 50 Duration: 3 Hours

(b) A discrete-time LTI system is described by the difference equation:

2. (a) Compare the computational complexity of 1024-point FFT (radix-2 Decimation-in-Frequency

Question 2 continues on the next page. . .

End of question paper. ALL THE BEST.

You might also like