Quiz 1 Answerak

Let's break down the questions one by one and work through their answers:
1. Explain the followings
(a) What is a zero-phase FIR filter? Suggest a suitable scheme for zero-phase filtering. Which types
of FIR digital filter can be converted into a zero-phase FIR filter and how?
A zero-phase FIR filter is a filter whose frequency response has no phase distortion, meaning the
phase response is zero at all frequencies. This is useful because it prevents distortion in the signal
due to phase shifts.
• Scheme for zero-phase filtering: A common approach to achieving zero-phase filtering is by

applying the filter forward and then backward. This is known as zero-phase filtering via
forward-backward filtering. In this approach, the signal is first filtered in the forward
direction, and then the filtered output is reversed and filtered again. This ensures that the
phase shift is compensated.
• FIR filters that can be converted to zero-phase FIR filters: FIR filters with linear phase can be
converted into zero-phase FIR filters. Linear phase FIR filters have a symmetric impulse
response, which ensures no phase distortion. These filters can include:
o Type I and Type II FIR filters: Symmetric impulse response.
o Type III and Type IV FIR filters: Anti-symmetric impulse response.
(b) Justify with a suitable example that the FIR filter with an even amplitude response will have a
linear phase if it has a symmetric impulse response.
For an FIR filter to have a linear phase, its impulse response must be either symmetric or anti-
symmetric. Let’s consider an example with a symmetric impulse response:
If the impulse response of the FIR filter is h(n)=h(N−1−n)h(n) = h(N - 1 - n)h(n)=h(N−1−n), where NNN
is the filter length, the phase response will be linear because the frequency response H(f)H(f)H(f) will
not introduce any nonlinear phase distortion.
Example: Let h(n)=[1,2,3,2,1]h(n) = [1, 2, 3, 2, 1]h(n)=[1,2,3,2,1]. This impulse response is symmetric.

The filter’s phase response will be linear, ensuring that all frequency components are delayed
equally, and no phase distortion is introduced.
2. Solve the followings
(a) Describe the process of designing an IFIR filter. What role does interpolation play, and how does
the filter achieve a reduction in computational complexity?
An Interpolated Finite Impulse Response (IFIR) filter is designed to reduce the computational
complexity of FIR filters by employing the concept of interpolation.
• Process:
1. Design a prototype filter: First, design a low-order prototype filter that meets the
specifications.
2. Interpolation: Apply interpolation by inserting zeros between the filter coefficients

of the prototype filter. This expands the bandwidth of the filter, thus decreasing the
computational load.
3. Band-limiting filter: After interpolation, a band-limiting filter is applied to ensure the

filter meets the original specifications.
• Role of interpolation: Interpolation increases the filter length but does so efficiently by
taking advantage of the zero-insertion technique, which doesn't significantly increase the
computational cost.
• Reduction in complexity: By using interpolation, fewer filter coefficients are needed

compared to a direct design of a filter with the same frequency characteristics, leading to
reduced multiplication operations during filtering.
(b) Assume H(z)H(z)H(z) is a causal stable minimum phase transfer function, and G(z)G(z)G(z) is a
causal stable transfer function with non-minimum phase but having the same magnitude
response. If h(n)h(n)h(n) and g(n)g(n)g(n) are impulse responses, then show that:
∣g(0)∣≤∣h(0)∣and∑i=0n∣g(i)∣2≤∑i=0n∣h(i)∣2|g(0)| \leq |h(0)| \quad \text{and} \quad \sum_{i=0}^{n}

|g(i)|^2 \leq \sum_{i=0}^{n} |h(i)|^2∣g(0)∣≤∣h(0)∣andi=0∑n∣g(i)∣2≤i=0∑n∣h(i)∣2
This part involves the analysis of minimum-phase and non-minimum phase systems. For minimum-
phase systems, the energy of the impulse response is concentrated at the start (i.e., lower indices),
while for non-minimum phase systems, it is spread out over time. Hence, we can show that the
energy in the first few samples of the minimum-phase system is larger than or equal to that in the
non-minimum-phase system.
3. Explain the followings
(a) Compare the performance of an IFIR filter to that of a FRM (Frequency Response Masking)
filter. How do the design approaches differ, and under what circumstances would one be preferred
over the other?
• IFIR Filter:
o Design approach: Based on interpolation of a low-order prototype filter. The primary

goal is to reduce the number of multiplications per output sample.
o Advantages: Reduced computational complexity compared to a standard FIR filter of

the same length.
o When to use: Best used when the application requires fewer computational
resources but can afford a slightly larger delay.
• FRM Filter:
o Design approach: Uses a combination of narrow-band and wide-band filters, where

a masking filter is applied to a prototype filter.
o Advantages: Offers a highly efficient design for very narrow-band or very wide-band
filters.
o When to use: Ideal for applications needing extreme selectivity with minimal
computational cost, such as communications systems where specific bands need to
be filtered out.
(b) Assume a LTI system with system function:
H(z)=1M∑i=0M−1Ai(z)H(z) = \frac{1}{M} \sum_{i=0}^{M-1} A_i(z)H(z)=M1i=0∑M−1Ai(z)
where Ai(z)A_i(z)Ai(z) are stable all-pass transfer functions. Show that H(z)H(z)H(z) is a BR (Bounded
Real) function.
• Bounded Real (BR) Function: A function is BR if it is stable and satisfies the condition
∣H(ejω)∣≤1|H(e^{j\omega})| \leq 1∣H(ejω)∣≤1 for all ω\omegaω. Since Ai(z)A_i(z)Ai(z) are all-
pass filters, they preserve the magnitude of the input, and because they are averaged, the
overall function will also be bounded.
(c) Show that it is always possible to design an FIR digital filter with exact linear phase with
suitable examples.
As discussed earlier, any FIR filter with a symmetric or anti-symmetric impulse response will have a
linear phase. For example, consider a filter with impulse response h(n)=[1,2,1]h(n) = [1, 2,
1]h(n)=[1,2,1]. This is symmetric, and therefore, the phase response will be linear

Quiz 1 Answerak

Uploaded by

Copyright:

Available Formats

Quiz 1 Answerak

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

Quiz 1 Answerak

Uploaded by

Copyright:

Available Formats

Let's break down the questions one by one and work through their answers:

1. Explain the followings

• Scheme for zero-phase filtering: A common approach to achieving zero-phase filtering is by

o Type I and Type II FIR filters: Symmetric impulse response.

o Type III and Type IV FIR filters: Anti-symmetric impulse response.

Example: Let h(n)=[1,2,3,2,1]h(n) = [1, 2, 3, 2, 1]h(n)=[1,2,3,2,1]. This impulse response is symmetric.

2. Interpolation: Apply interpolation by inserting zeros between the filter coefficients

3. Band-limiting filter: After interpolation, a band-limiting filter is applied to ensure the

• Reduction in complexity: By using interpolation, fewer filter coefficients are needed

∣g(0)∣≤∣h(0)∣and∑i=0n∣g(i)∣2≤∑i=0n∣h(i)∣2|g(0)| \leq |h(0)| \quad \text{and} \quad \sum_{i=0}^{n}

o Design approach: Based on interpolation of a low-order prototype filter. The primary

o Advantages: Reduced computational complexity compared to a standard FIR filter of

o Design approach: Uses a combination of narrow-band and wide-band filters, where

(b) Assume a LTI system with system function:

H(z)=1M∑i=0M−1Ai(z)H(z) = \frac{1}{M} \sum_{i=0}^{M-1} A_i(z)H(z)=M1i=0∑M−1Ai(z)

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.