Experiment 3 Filters II Filter Design with MATLAB

The objective of this experiment is to gain some experience in designing filters with desired specifications. You will work with a number of tools helping you in designing filters in MATLAB.

Prelab P1. Review of Filters Frequency Domain

The effect of a system on a signal input to it can be described in both time and frequency domains. We observed the time representation of the system in last experiment. Here we will study the frequency domain representation of the system as a filter. Different frequency components of the input signal experience different attenuation/amplification and delays as they pass through the filter thus a filter alters the signal input to it in frequency domain (Fig. P1). One of the main applications of a filter is to filter out the unwanted parts of an input signal. The unwanted part might be noise present out of the frequency band occupied by the desired signal (high frequency noise) or other interfering signals that are unwanted, e.g. when you tune to a radio station, the radio receiver will filter out other stations so you would hear a clear sound (Fig. P2).

Figure P1 The alteration of frequency components of a signal input to a filter Simple filters can be categorized into four major categories: lowpass, highpass, bandpass and bandstop (or band-reject). Figure P3 shows the ideal forms of these four filters in the frequency domain. As the name suggests a lowpass filter for instance, passes the low frequency components of a signal while blocking the high frequency ones. As you have learned the convolution operation in time domain is equivalent to multiplication in frequency domain. Thus, multiplying the frequency domain representation of the input signal with that of the filter, high frequency components of the input signal will be multiplied in zero and would thus vanish, while the lower frequency components will be multiplied by one and would pass without any change.

Noise

f Interfering stations

Desired station

Figure P2 Some basic applications of filters: top) noise reduction by removing out of band (high frequency noise), bottom) removing interfering signals: passing of a tuned channel in a radio. In this experiment you will learn how to use some tools in MATLAB signal processing toolbox to design filters with your desired characteristics.

Figure P3 The ideal frequency response of the four basic filter types: 1) lowpass, 2) highpass, 3) bandpass and 4) bandstop (or band-reject)

P2. Filter Design

Before designing a filter, you should know what parameters need to be known for the design. The ideal filters as shown in figure P3, cannot be physically realized, since they are noncausal, i.e. to generate the output at instance n in time, they require the knowledge of future samples of the input signal, i.e. at n + 1, n + 2,... , which is impossible. The design techniques are as a result, approximations of the ideal case. The digital filter design methods fall into two main categories, FIR filter design and IIR filter design. You will learn more about these types in your future courses. In this experiment you will be introduced to some tools for basic IIR filter design. First of all you should know the parameters of the filter that you are going to design. Some of these parameters are described in this section. dB Notation Something you will very often encounter when dealing with filters is the deci-Bell (dB) notation. The reason is that when youre dealing with a wide range of variation in the magnitude of a quantity, say 5 orders of magnitude difference between maximum and minimum, it is hard to work with the numbers. If you use the logarithm of the magnitude, the variations will be too small! So Alexander Graham Bell, invented this new notation named after him by using 20 times the logarithm of the magnitude. Thus
x dB = 20 log 10 x x = 10 x dB
20

Some figures that you should know without using your calculator are 3 dB and its multiples. 3 dB is almost equal to 2, thus if you have a quantity that is 3 dB greater than another, it is actually two times greater than it. (Note that you are using logarithm, thus addition of the logarithms is equivalent to multiplication of the numbers themselves.) By analogy, -3 dB, 6 dB and -6 dB are equivalent to 0.5, 4 and 0.25 respectively. Passband Parameters The passband of the ideal filter is the band of frequencies where the filter frequency response is not zero. If a signals frequency content lies in this band, the signal will pass without any alteration. In case of practical filters, the frequency response does not have an abrupt change from zero to one or the value in passband (Fig P4). Instead, there is a transition from a higher level to a lower level. Hence, for a practical filter the band of frequencies over which the magnitude of frequency response is greater than a threshold is called the passband. This threshold is usually set at -3 dB of the maximum magnitude in passband. The frequency at which the filters frequency response magnitude reaches this threshold is called the cut-off frequency of the filter ( f c ) . If the frequency response does not decrease monotonically in the passband the filter is said to have ripples in the passband. The difference between the maximum and minimum magnitude values in the passband defines the passband ripple factor R p usually given in dB. The maximum value of the magnitude in passband is called the gain of the filter. Stopband Parameters The stopband of the ideal filter is the band of frequencies where the filter frequency response is zero. Thus the frequency components of the signal lying in this band will not

pass and will be fully removed. Again, since the transition from high to low is not abrupt in practical filters, a threshold level defines this region. So, the band of frequencies, where the magnitude of the frequency response of the filter drops below a threshold level is called stopband. The threshold value is dependent on the application of the filter usually about 40 dB and is usually denoted by As or Rs . If there are ripples in the stopband, the maximum value of the magnitude of frequency response in the stopband shall lie below the threshold. The frequency at which the magnitude of the filter response first falls below the threshold value is called the stopband frequency denoted by f s .

Figure P4 The filter parameters: 1) passband, 2) stopbands, 3) cut-off frequencies, 4) stopband frequencies, 5) passband ripple, 6) stopband level, 7) filter gain.

P3. Four Classical IIR Filter Types

In this section I will introduce and compare the four classical IIR filter types, namely Butterworth, Chebychev Type I and II, and Elliptic filters. You will encounter these filters and probably their design methodologies in your future courses. Since straightforward algorithms for design and implementation of these filters exist, many software programs help you in designing these filters by simply specifying their parameters. Hence, for the sake of this experiment you need not know all the mathematical detail involved in the design of these filters1. All filters here refer to the lowpass case. You can simply generalize the results to the other types by analogy.
If you are interested in further study of these details you may refer to Digital Signal Processing, System Analysis and Design, by P.S.R Diniz, E.A.B. da Silva and S.L. Netto, Cambridge University Press, 2002
1

Butterworth Filter The Butterworth filter of order N , also called the maximally flat filter, is an approximation of the ideal filter, which the first 2 N 1 derivatives of its magnitude squared are zero. As a result the frequency response of this filter decreases monotonically with frequency and H ( f = f c ) = 1 2 . The decrease is very slow in the passband and quick in the stopband. In a design problem where no ripple is acceptable in passband and stopband, Butterworth filter is a good choice.

Figure P5 The frequency response of a Butterworth filter with c = 1 Chebychev Type I Filter A Chebychev type I approximates the ideal filter by minimizing the absolute difference between the resulting filter and the ideal filter, over the passband. This filter results in appearance of ripples in the passband. The ripples have a constant magnitude of R p which is one of the design parameters of this filter. The stopband response is maximally flat. The transition from passband to stopband occurs faster than the Butterworth filter: H ( f = f c ) = R p . If the ripples in the passband are acceptable, a Chebychev filter usually require a lower-order transfer function than a Butterworth filter for the same specifications.

Figure P6 The frequency response of a Chebychev type I filter with c = 1

Chebychev Type II Filter A Chebychev type I approximates the ideal filter by minimizing the absolute difference between the resulting filter and the ideal filter, over the stopband. This filter results in appearance of ripples in the stopband. The ripples have a constant magnitude of Rs which is one of the design parameters of this filter. The passband response is maximally flat and H ( f = f c ) = Rs . The transition from passband to stopband occurs slower than the type I filter and for even filter order, it never reaches zero. Its advantage over type I is the absence of ripples in the passband.

Figure P7 The frequency response of a Chebychev type II filter with s = 1 Elliptic Filter The elliptic filter results in the steepest transition from passband to stopband by allowing ripples in both passband and stopband. This filter usually satisfies the desired filter specifications with the lowest order among the four types studied. The passband and stopband ripples R p and Rs are also required as design parameters.

Figure P8 The frequency response of an elliptic filter with c = 1

MATLAB sptool: Filter Design and Signal Processing Tool

1. Adding Phone Effect to a Sound In this section you will learn how to use a powerful signal processing tool in MATLAB, sptool which is part of the MATLABs signal processing toolbox and helps you with filter design and signal processing with an easy to use GUI. In what follows, you will design a bandpass filter that resembles the bandpass filter in a telephone system. You will then apply this filter to the French speaker sound file you used before, to create a phone effect to the sound. At MATLAB command prompt type sptool. In the Filters list click on the New button. This creates a new filter design. In the sampling frequency edit box enter 11025 as the sampling frequency. From the algorithm dropbox, select Butterworth IIR filter. Select bandpass in the filter type dropbox. Enter 200 and 1200 for the low and high stopband frequencies respectively. Enter 250 and 1000 for the low and high passband frequencies respectively. Click the Apply button at the bottom of the left panel. Close the filter design tool. Your filter is now designed and ready for use! You should import the French speakers sound file into sptool, apply your filter to it: At MATLAB command prompt load the .wav file french.wav. Downsample the sound from 44100 to 11025Hz (dont forget to update FS). In sptool, in File menu select Import Set Import as to signal. Select from workspace and select the name of the vector containing the sound data and click the right arrow to import the data Select the sampling frequency from the workspace variables and import it. Enter a name for the imported signal, the default is sig1. Click OK to complete import. From the Filters list select your filter design and click the apply button. Leave the defaults as they are and click OK. In the Signals list, you will see the result of filtering, select it, hold down control key and select the input signal. Click the View button. If you cant see both signals, click on the select trace output signal. in the toolbar and select the

Click the play sound in the toolbar to listen to the output. Switch the selected trace from output to input, play the signals and note the difference. Now go back to filter design screen to compare different filter design methods with each other. Make sure the value of R p is set at 3 dB and Rs at 20 dB.

Note the shape of the Butterworth filter response and specifically the stopband frequencies. Now change the filter type to Chebychev type I. Take note of the actual stopband frequencies and the shape of the filter response. Repeat for the Chebychev type II and elliptic filter types. What is your deduction?

To view the impulse response of each filter, go back to sptool main window and select your filter and click the View button. A new tool window appears. You may see the phase and amplitude response of the filter you have designed in this window. From the left panel, check the Impulse Response. Use the zoom tool , to zoom in and see the first portion

of the response. Use the zoom all tool , to select another potion or use the zoom tool to zoom in again. Uncheck the Magnitude and Phase boxes and check the Zeros and Poles. Zoom out to see the full plot. In the design window, change the filter type. What happens to the position of the poles and zeros? Can you compare the four basic designs? 2. Out of Band Noise Removal In this part you will use a lowpass filter to remove the high frequency noise in a signal. At MATLAB command prompt use size command to determine the number of samples in the downsampled French speaker sound you loaded before. Generate a random noise vector n with the same number of samples as the French speakers signal. Use MATLAB randn command to generate the noise vector whose samples have Gaussian probability distribution. Scale down the noise by multiplying your vector by 0.1. Add the noise and your original sound to make a new vector. Make sure you dont destroy your original sound file. Import the noise and the noisy sound in sptool. View the noise, noisy sound and the original sound using View button. Listen to the three signals. Now design a lowpass filter with following specifications: o f c = 1500 Hz and f s = 1650 Hz. o No ripple in passband is allowed. o Your filter order should not be more than 15. (Check the right panel to see the order of your filter) o Rs should be at least 30 dB. Apply your filter to the noise signal and to the noise. View the results and signals before filtering. Listen to them and compare them. Now try to export your filter design to MATLAB: In sptool main window in File menu select Export Select your filter name in the left panel To see the transfer function of the filter, check the Export Filter as TF Object.

Click the Export to worksapce button. At MATLAB command prompt, type the name of your filter design to see its transfer function. You will see the name of your filter in MATLABs workspace panel. Return to sptool and again export your filter design now with the Export Filter as TF Object unchecked. Confirm replacement of your old workspace filter export. At MATLAB command prompt, type the name of your filter you should see something like this:
filt2 = tf: ss: zpk: sos: imp: step: t: H: G: f: specs: Fs: type: lineinfo: SPTIdentifier: label: [1x1 struct] [] [1x1 struct] [] [] [] [] [1x4096 double] [4096x1 double] [1x4096 double] [1x1 struct] 11025 'design' [1x1 struct] [1x1 struct] 'filt2'

You will see the name of your filter instead of filt2. To access the transfer function parameters of your filter use this syntax and replace filt2 with your filters name: filt2.tf.num to see the numerator coefficients of your filter and filt2.tf.den to see its denominator coefficients. You may access the other parameters of the filter, like location of poles and zeros and sampling frequency in the same way.