Scribd
Upload
14 views

Digital Filters

Uploaded by

rjchumacera
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
14 views

Digital Filters

Uploaded by

rjchumacera
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 42

Manuel S.

Enverga University Foundation


Lucena City, Philippines
Granted Autonomous Status
CHED CEB Res. 076-2009

Filtering
•
Part IV
Impulse Response and Tap Coefficient
• FIR Filter Analysis and Design
• FIR Band Pass, High pass and Band Stop Filters

Sherwin C. Lagrama.,RECE,MMEM
College Of Engineering and Technical Department
1st Sem 2021-2022

COLLEGE OF ENGINEERING AND TECHNICAL DEPARTMENT


Telefax No. (042) 710-3151; e-mail:[email protected]
is a device or process that removes some unwanted
component or feature from a signal. Filtering is a
class of signal processing, the defining feature of filters
being the complete or partial suppression of some aspect
of the signal.

Most often, this means removing some frequencies


and not others in order to suppress interfering signals
and reduce background noise.
Introduction
Basic Types of Filters:

a) Low Pass Filter

b) High Pass Filter

c) Band Pass Filter

d) Band Stop Filter


Low Pass Filter
Low Pass Filter

-it allows signal frequency below the low cut-off


frequency to pass and stops frequencies above the cut-
off frequency. Its is commonly used to help reduce
environmental noise and provide a smoother signal. Its
main function is to remove high frequency interference
noise.
Low Pass Filter

Magnitude Response of a normalized low pass filter


High Pass Filter
High Pass Filter

-it allows frequencies higher than the cut-off to


pass and removes any steady direct current component
or slow fluctuations from the signal.

-it is used to eliminate low frequencies which are


produced by slow changes, such as conductivity shift
within a material, varying distance to an edge while
scanning parallel to it, or out-of-round holes in fastener
hole inspection.
High Pass Filter

Magnitude Response of a Normalized High Pass Filter


LPF and HPF

Comparison Chart for LPF


and HPF
Band Pass and Band Stop Filters
Band Pass Filters:

-is a device that passes frequencies within a


certain range and rejects (attenuates) frequencies outside
that range.

Band Stop Filter

-is a filter that passes most frequencies unaltered,


but attenuates those in a specific range to very low levels.
It is the opposite of a band-pass filter
Band Pass Filter

Magnitude Response of a Normalized Band Pass Filter


Band Stop Filter

Magnitude Response of a Normalized Band Stop Filter

h(n) = hlp + hhp


Original Signal
Example of Signal Filtering Outputs

Original Signal
Filtered Signals
Low Pass High Pass
Filter Filter
Filtered Signals

Band Stop Filter


Band Pass Filter
Window Design Method
Window Technique Steps:
Characteristics of different windowing functions

TABLE 4.1

Window Type Peak approximation Transition


error Bandwidth

Rectangular -21

Bartlett -25

Hanning -44

Hamming -53

Blackman -74
Shapes of Window Functions

o line = Bartlett
dashed line = Blackman

+ line = Hanning
solid line = hamming
Window Type Definition

Rectangular
Iif N is odd:

Bartlett If N is even:
Window Type Definition

Hanning

Hamming

Blackman
1. Solved Problem
Example:
The low pass filter must be
designed according
to the following specifications:

NOTE:
1. Solved Problem
1. Solved Problem
1. Solved Problem
Therefore, the number of tap coefficients required
for transition band of ∆ω is N=M+1=55 taps
1. Solved Problem
1. Solved Problem

0.00018

0.000373
1. Solved Problem
1. Solved Problem

when all the tap coefficients are computed, we can


observe that filter coefficients are symmetric, that is:

This is the characteristic of a linear-phase FIR filter.


The table shown summarizes the calculation result for the
filter.
Tap Coefficients of the filter

h(0) = h(54) 0.00018 h(14) = h(40) 0.0053


h(1) = h(53) -0.000373 h(15) = h(39) 0.0164
h(2) = h(52) -0.0011 h(16) = h(38) 0.0074
h(3) = h(51) 0.0000 h(17) = h(37) -0.0163
h(4) = h(50) 0.0016 h(18) = h(36) -0.0252
h(5) = h(49) 0.0016 h(19) = h(35) 0.0000
h(6) = h(48) -0.0011 h(20) = h(34) 0.0359
h(7) = h(47) -0.0036 h(21) = h(33) 0.0335
h(8) = h(46) -0.0017 h(22) = h(32) -0.0225
h(9) = h(45) 0.0039 h(23) = h(31) -0.0757
h(10) = h(44) 0.0062 h(24) = h(30) -0.0359
h(11) = h(43) -0.0000 h(25) = h(29) 0.1111
h(12) = h(42) 0.0090 h(26) = h(28) 0.2932
h(13) = h(41) -0.0083 h(27) 0.375
1. Solved Problem

We can further analyze the designed filter utilizing


what we have learned in earlier chapters . For instance,
if we obtain the impulse response of the filter, as shown
in Fig 4.7, we can observe the symmetric nature of the FIR
filter clearly. Similarly, we can study its frequency
response and verify the filter specifications and
characteristics. Fig 4.8 (a) shows the magnitude response
(filter gain) of the filter. In Fig 4.8 (b) the same gain
plotted in dBs for clarity and Fig 4.8 (c) shows the phase
characteristic of the filter.
1. Solved Problem
Impulse Response
2. Solved Problem
A linear-phase FIR filter with the following specification
is to be designed using the windowing technique:
• Sampling Frequency=16 kHz
• Pass band: 0 to 1.2 kHz
• Stop band: 4 kHz to 8 kHz
• Peak Approximation Error, δ=0.0125
a) Determine the centre of transition band
b) Determine the type of window to be used and number
of filter coefficients needed.
c) Calculate the filter coefficients for n=10,11,12,13
(n=0,..,N-1)
2. Solved Problem
Solution:
a. Pass band:

Stop band:

Width of transition band:


2. Solved Problem
b. Peak Approximation Error

Therefore, Hanning Window with -44 is chosen.


2. Solved Problem
c.

But
Wc=0.325π
M=24 23
23 23
2. Solved Problem
23
23 23
0.2039678

23

23 23
0.3097

23
23 23

0.325

23
XXXXXXXX
23 23
Summary of Filter Equations:

a. Lowpass Filter

b. High Pass Filter

c. Band Pass Filter

d. Band Stop Filter h(n) = hLP + hHP


1. All research(s) that returned to students including solved
problems with remarks should be corrected. (e.g cover
page, short bond paper, standard format). And submit on
October 23, 2011
2. The Lab report should be submitted next meeting and no
extension will be given.
3. All students who are not submitted their assignments will
be given last consideration next meeting to accomplish all
pending documents. (no extension)
4. Next meeting, all documents must be submitted using
portfolio folder compile orderly with paper fastener.
5. Use sticker paper to label your portfolio. (Instructor will
STICKER PAPER
“not glossy”
• White background

Lagrama, Sherwin C.
Signal Spectra Lec
Sun 2pm-4pm
• Q4-3 Design a low pass FIR filter with following specifications:
• Sampling frequency: 15 kHz
• Pass band: 0 to 2 kHz
• Peak approximation error (δ): 0.002
• Filter length (N): 31
• To strictly meet the specifications, what window should be used in the design?
• What is the stop band of the filter?
• Calculate the tap coefficients h(14) and h(15) of the filter.
• Q4-4 A 33-tap FIR low pass filter is designed using the windowing technique. Hanning window is
used in the design. Some of the coefficients of the filter are found to be:
• h(15) = 0.29125
• h(16) = 0.375
• h(17) = 0.29125
• If the sampling frequency used is 8 kHz, determine
• the pass band and stop band frequency ranges in Hz.
• the minimum stop band attenuation in dB.
• What is the benefit in using the Hanning window as compared to the use of rectangular window?
What is the disadvantage?
Thank you
Instructional materials by
Engr. SCLagrama.,RECE,MMEM

You might also like