Basic Questions on FIR Filters
Q1: What is an FIR filter?
A: FIR stands for Finite Impulse Response filter. It is a type of digital filter with a finite duration
impulse response, which settles to zero in finite time.
Q2: What are the advantages of FIR filters?
A: FIR filters are always stable, have linear phase response (important for signal fidelity), and
can be easily implemented.
Q3: What is the difference between FIR and IIR filters?
A: FIR filters have finite duration impulse response and are inherently stable. IIR (Infinite
Impulse Response) filters have feedback and can be unstable but require fewer coefficients for
sharp filters.
🔹 Windowing Method in FIR Filter Design
Q4: What is the purpose of using a window function in FIR filter design?
A: A window function is used to truncate the ideal infinite impulse response to a finite length. It
helps control the trade-off between the main lobe width and side lobe levels in the frequency
response.
Q5: What is a Hamming window?
A: The Hamming window is a type of window function defined as
w(n) = 0.54 - 0.46 cos(2πn / (N - 1))
It reduces the side lobes better than a rectangular window, offering better stopband attenuation.
Q6: What is a rectangular window?
A: The rectangular window simply truncates the impulse response without any smoothing. It is
defined as
w(n) = 1 for all n.
It has the narrowest main lobe but the highest side lobes, leading to poorer stopband
attenuation.
🔹 Filter Types
Q7: How is a low-pass FIR filter designed using the window method?
A: First, derive the ideal impulse response of the low-pass filter, then multiply it with a chosen
window (e.g., Hamming or rectangular) to get the FIR filter coefficients.
Q8: What changes when designing a high-pass filter instead of a low-pass filter?
A: The impulse response is derived differently using spectral inversion or shifting. The ideal
high-pass impulse response is obtained by subtracting the low-pass response from a delta
function.
Q9: What is the difference between a bandpass and a bandstop filter?
A: A bandpass filter allows a range of frequencies to pass and attenuates others. A bandstop
(or notch) filter does the opposite—it attenuates a range of frequencies while allowing others to
pass.
Q10: How is a bandpass FIR filter designed using the window method?
A: Use the difference of two low-pass filters to create a bandpass filter: h[n] = h_lp2[n] -
h_lp1[n], where h_lp2 has a higher cutoff than h_lp1. Then apply the window function.
🔹 Window Comparison in Filter Design
Q11: Why is the Hamming window preferred over the rectangular window?
A: Because the Hamming window has lower side lobes, leading to better stopband attenuation
and less spectral leakage.
Q12: What are the drawbacks of using a rectangular window in FIR filter design?
A: It has high side lobe levels, which result in poor stopband performance and higher ripples in
the frequency response.
🔹 Application-Specific Questions
Q13: What applications use FIR low-pass filters?
A: Audio signal smoothing, anti-aliasing filters, and removing high-frequency noise.
Q14: Where is a high-pass FIR filter used?
A: In applications like speech processing, removing low-frequency hum, or highlighting rapid
changes in signals.
Q15: What are some uses of FIR bandstop filters?
A: Removing narrowband interference like 60 Hz power line noise from biomedical or audio
signals.
Q16: What happens to the transition bandwidth when using a Hamming window?
A: The transition band is wider than with more aggressive windows (like Blackman), but better
than with a rectangular window.
🔹 Practical Implementation Questions
Q17: How do you choose the length of the FIR filter?
A: It depends on desired transition width and stopband attenuation. Longer filters provide better
frequency selectivity.
Q18: What is the effect of increasing the filter length (N)?
A: It narrows the transition band and improves frequency resolution but increases
computational complexity.
Q19: Why must FIR filters be linear phase in some applications?
A: Linear phase ensures that all frequency components of a signal are delayed equally,
preserving the waveform shape—critical in audio and communication systems.
1.) Define Convolution
A: Convolution is a mathematical operation that combines two signals to produce a third signal,
showing how one signal modifies the other.
2.) What are the types of Convolution?
A:
● Linear Convolution
● Circular Convolution
3.) What is Linear Convolution & Circular Convolution?
A:
● Linear Convolution: Combines two signals to get an output of length (N + M - 1); used
in LTI systems.
● Circular Convolution: Assumes periodic signals; output has the same length as inputs.
Common in DFT/FFT.
4.) State the methods available to compute Convolution Sum
A:
● Graphical method (flip and slide)
● Matrix method
● Tabular method
● Using Z-transform
● Using DFT (for circular convolution)
5.) What is the significance of convolution?
A:
It helps analyze LTI systems, determine output for any input, and is essential in filtering, signal
processing, and system analysis.
. What is the difference between DTFT and DFT?
A:
● DTFT (Discrete-Time Fourier Transform) gives a continuous spectrum for a
discrete-time signal; it is theoretical and not directly computable.
● DFT (Discrete Fourier Transform) gives a discrete spectrum and is computable,
used in practical implementations (via FFT).
2. Write any Two Properties of DFT
A:
● Linearity: DFT of a sum is the sum of the DFTs.
● Circular Convolution: DFT converts circular convolution in time domain to multiplication
in frequency domain.
3. What is Zero Padding and its Effect on Magnitude Spectrum?
A:
● Zero Padding: Adding zeros to a signal to increase its length before computing the DFT.
● Effect: Increases frequency resolution; gives a smoother and more detailed magnitude
spectrum, but does not add new information.
4. Write Two Properties of Twiddle Factor (W = e^(-j2πk/N))
A:
● Periodicity: WNk+N=WNkW_N^{k+N} = W_N^k
● Symmetry: WNN−k=WN−kW_N^{N-k} = W_N^{-k}
●
