(SEM V) THEORY EXAMINATION 2020-21 DIGITAL SIGNAL PROCESSING

SECTION A (Any 2 Questions Explained Briefly)
 

What is Gibb’s phenomenon in FIR filters?
Gibb’s phenomenon refers to the oscillations that appear near sharp discontinuities in the frequency response of FIR filters designed using Fourier series methods. These ripples do not completely disappear even if the filter length is increased, though their width reduces. Windowing techniques are used to reduce this effect.
 

Distinguish between recursive and non-recursive digital filters.
Recursive digital filters use feedback, meaning the output depends on present as well as past outputs, and they are also known as IIR filters. Non-recursive digital filters do not use feedback and depend only on present and past inputs; these are called FIR filters. Recursive filters are more efficient but may be unstable, while non-recursive filters are always stable.
 

SECTION B (Any 2 Questions Explained)
 

What are the advantages of FFT over DFT? Explain DIT algorithm.
The Fast Fourier Transform significantly reduces the number of computations required to evaluate the Discrete Fourier Transform. While DFT requires O(N²) operations, FFT reduces it to O(N log N), making it much faster for large data sets. The Decimation-In-Time (DIT) FFT algorithm works by dividing the input sequence into smaller subsequences and recursively computing their DFTs, followed by combining the results.
 

Explain multirate signal processing and its advantages.
Multirate signal processing involves changing the sampling rate of a signal using operations such as up-sampling and down-sampling. It is used to reduce computational complexity, improve efficiency, and design flexible digital filters. Multirate techniques are widely used in speech processing, audio coding, and communication systems.
 

SECTION C (Any 2 Questions Explained)
 

Explain direct form-I and direct form-II realization of a digital system.
Direct form-I realization implements the difference equation using separate delay elements for input and output, resulting in more memory usage. Direct form-II combines these delays and uses minimum memory by sharing delay elements. Both structures represent the same system, but direct form-II is more efficient in terms of storage.
 

Explain finite word length effects in digital filters.
Finite word length effects occur due to limited precision in representing numbers in digital systems. These effects include coefficient quantization error and round-off noise, which can distort the output and affect filter stability. Proper scaling and word-length selection help minimize these errors.
 

Most Questions in This PDF Are Related To
 

Most questions in the Digital Signal Processing (KEC-503) paper are related to DFT and FFT algorithms, digital filter design (FIR and IIR), frequency transformations, realization structures, multirate signal processing, finite word length effects, convolution, and spectral analysis. The paper mainly focuses on analysis, design, and implementation of digital signal processing systems.