Digital Signal Processing with

LabVIEW 8.6

Lab # 09 and 10

Digital filtering using Digital Signal

Processing

Adnan Niazi

Lab Engineer

Signal Processing/Signal & Systems Lab

CECOS University of IT & Emerging Sciences

10th- 11thApril 2009

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Agenda

▌ Introduction to filtering

▌ Analog Filtering

▌ Digital Filtering

▌ What is FIR filter

▌ What is IIR Filter

▌ Types of FIR and IIR filters

▌ What is group delay and why its important

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Filtering

▌ Filtering removes unwanted signals from the wanted

ones

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of Filtering

▌ Analog Filtering

▌ Digital Filtering

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Analog Filtering

▌ Analog filters work on signals in continuous time

domain

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of Analog Filters

▌ Analog filters are of three types:

 Passive Filters

• Made from Resistors, Capacitors and Inductors

 Active Filters

• Made from Opamps, Resistors and Capacitors

 Switched Capacitor Filters

• Made from Switched capacitors

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Analog Filter Topologies

▌ Butterworth Filter

▌ Bessel Filter

▌ Besselworth Filter

▌ Chebychev Filter

▌ Inverse Chebychev Filter

▌ Elliptic and Cauer Filter

▌ Gaussian 6dB Filter

▌ Gaussian 12dB Filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Why Digital filters better than Analog Filtering

▌ Analog filters become unstable when the order is

increased beyond 10

▌ This means that you can’t make filters with very steep

roll offs

▌ Digital filters are more accurate where as analog filters

are susceptible to variation in their response due to drift

and tolerance in component values

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Digital Filter

▌ Digital filters work on

samples in discrete time

domain

▌ Digital filtering can be

carried out on a

 DSP

 FPGA

 & to some extent on a

microcontroller

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Advantages of Digital Filtering

▌ Digital filters are very accurate as there is no component

drift

▌ Digital filters can achieve much sharper roll off that is

impossible to obtain with analog filters

▌ FIR filters are unconditionally stable whereas an analog

filter may become unstable if designed improperly

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Some Filter Lingo

▌ Transition Band or Skirt

 The range of frequencies between Passband and

Stopband

▌ Roll Off

 Roll Off refers to transition of from Passband to Stopband.

The higher the roll off the smaller the transition band

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of Filter Response

There are four major types of filter responses

▌ Low Pass filter

▌ High Pass filter

▌ Band Pass filter

▌ Band stop filter

 Notch filter: is a special type of band stop filter where the

stop band is very narrow

▌ Comb Filter

 Comb filter has notches at the harmonics as well

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of Filter Responses

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Special Filter Responses

▌ Notch Filter

 Notch filter is special type of band stop filter in which the

stop band is very narrow

 Notch filters are mostly used to remove power line

interference (50-60Hz) from the signals

▌ Comb Filter

 Comb filter has notches at multiples of a frequency

 Comb filter is used to filter out a frequency along with all

its harmonics

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Comb Filter applied to ECG

Time Domain plot of ECG Signal

Frequency Domain Plot of ECG

with 60Hz noise and its

Signal with 60Hz noise and its

harmonics noise

Magnitude response of the

Filtered vs. ideal ECG signal

Comb Filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Magnitude Response Parameters

Pass band

Ripple

Stop band

ripple

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Filter Parameters

▌ Sample rate

 Specifies the sample rate of the signal on which filtering is

to be performed

▌ Passband Edge Frequency

 Specifies the corner frequency of the filter. Frequency

beyond this frequency will be filtered out. Its also referred

to as the corner frequency

▌ Stopband Edge Frequency

 Specifies the frequency after which there is no roll off

▌ Passband Ripple

 Specifies how much ripple should in the Passband

frequencies

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Filter Parameters

▌ Passband Ripple

 Specifies how much ripple should in the stop band

frequencies

▌ Stop Band Attenuation

 Specifies how much the Stopband frequencies should be

attenuated before stop band ripples start to appear

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of Digital Filters

▌ There are two types of digital filters

 FIR Filter

 IIR Filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

FIR Filter

▌ FIR stands for Finite Impulse Response Filters

▌ The impulse response of FIR Filter decays after a finite

amount of time

▌ The mathematical equation for FIR filter is

 Where a’s represent the Feed forward Coefficient

▌ Coefficients are also called weights

▌ It is weights that give a filter desired properties

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Finite Impulse Response (FIR)

▌ FIR filters are called so because their impulse response

has a finite duration.

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of FIR Filter

▌ Equi-Ripple Filter

▌ Parks - McClellan Filter

▌ Savitzky-Golay Smoothing Filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

IIR Filter

▌ IIR stands for Infinite Impulse Response

▌ IIR filters have an impulse response that extends to

infinity

▌ The equation for IIR Filters is

 Where a’s represent the Feed forward Coefficient and

b’s represent the Feed Back Coefficients

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Infinite Impulse Response

▌ IIR filters Impulse Reponses never completely dies out

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Types of IIR Filter

▌ Butterworth Filter

▌ Bessel Filter

▌ Chebychev Filter

▌ Inverse Chebychev Filter

▌ Kaiser Window Filter

▌ Dolph- Chebychev Filter

▌ IIR smoothing Filter

▌ Least Path Norm Filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Advantage of FIR Filter

▌ Filters that are normally impossible to make with IIR

filters can be made with FIR filters

▌ There is no analog counterpart for the FIR Filters

▌ FIR filters have linear phase or constant group delay

▌ FIR filters are unconditionally stable (BIBO) as they have

no poles in their transfer function.

▌ Fixed point Implementation is easier to implement

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Advantages of IIR filter

▌ IIR filters consume less memory and DSP resources

▌ IIR filters can become unstable if designed improperly

▌ IIR filters suffer from a phenomenon called Ringing. The

effect of a wrong input sample is present for a very long

time because of the infinite impulse response of the IIR

filter

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Disadvantage of IIR Filter

▌ Difficult to implement on fixed point processor

▌ IIR filters have variable group delay which result in the

signal distortion

Contact: [email protected]

CECOS University of IT & Emerging Sciences

What is Group delay

▌ Group delay is the rate of change of the phase of the

filters frequency response

▌ The frequency reposes of a filter is given by:

▌ Group delay is a measure of the phase distortion of a

signal after it is passed through the filter

▌ Ideally group delay must be constant

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Group delay Demo

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Linearity of the Filter

▌ A filter is linear if it has constant group delay or Linear

Phase delay or phase response

▌ Linearity states that all the signals passing through the

signal must suffer the same time delay

▌ In other words higher frequencies should suffer more

delay and lower frequencies should suffer smaller

delays

▌ If a filter has constant group delay or linear phase delay,

the passed signal suffer from minimal distortion

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Task 1

▌ Generate a 1000Hz Embedded Sine wave of amplitude 1

sampled at 36KHz with a DC offset of 0.5

 Display the time domain Signal on the front Panel

 Display the spectrum of the Signal. The spectrum should

be from 0 to fs/2.

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Task 2

▌ Design a Filter that could remove the DC offset from the

embedded signal discussed in Task 1

▌ The filter pass band should be 100Hz

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Lab Report for Task 1

▌ The lab Report must contain

 A picture of the front panel

 A picture of the block diagram

 Always Write Full Question before beginning to answer it

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Lab Report for Task 2

▌ The lab report must contain

 Picture of the front panel

 Picture of the block diagram

 Always Write Full Question before beginning to answer it

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Final Test for Section A

▌ Section A test will on Monday 4th of May 2009

▌ From 9:00 AM to 9:30 AM students will attempt 30 MCQ

questions

▌ From 9:45 AM to 10:15 AM student will perform a

practical on the PCs

▌ The practical performance will be related to all that you

have learned in the lab.

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Final Test for Section D

▌ Section D test will on Tuesday 5th of May 2009

▌ From 9:00 AM to 9:30 AM students will attempt 30 MCQ

questions

▌ From 9:45 AM to 10:15 AM student will perform a

practical on the PCs

▌ The practical performance will be related to all that you

have learned in the lab.

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Do you Need to worry?

▌ If you have listened to me in the class then there will be

no problem attempting the MCQs

▌ If you have done all the lab reports yourself then there

will be no problem in practical performance

Contact: [email protected]

CECOS University of IT & Emerging Sciences

Automated Projector Calibration

Contact: [email protected]

CECOS University of IT & Emerging Sciences