LAB 5: Introduction to Simulink and

Filter Design using MATLAB.

Objective
This lab:

 Gives you a deeper understanding of the analog and digital filter design techniques in
MATLAB.
 Gives you a deeper understanding of the filter design techniques in MATLAB using the
Filter Design & Analysis Tool (FDAT).
 Helps you to represent, play, construct and plot audio signals in MATLAB.
 Introduces you to Simulink.

Hint: Part of this lab uses Simulink, a companion to MATLAB. The lab is self-contained, in the
sense that no additional documentation for Simulink is needed.

Introduction
 Lowpass Filter: The lowpass filter (LP-filter) is characterized by the attenuation of the
higher frequencies and passing the low, sometimes with a gain. The phase characteristics
of this filter depend on the order of the filter and the implementation [1].
 Highpass filters: The Highpass filter (HP-filter) is characterized by the attenuation of
the low frequencies and passing the high, sometimes with a gain. The phase
characteristics of this filter depend on the order of the filter and the implementation [1].
 Bandpass Filter: The bandpass filter selectively allows passage for all the frequencies of
the certain range and rejects/attenuates frequencies out of that range, sometimes with a

Page 1 of 26
 gain. The phase characteristics of this filter depend on the order of the filter and the
implementation [1].
 Bandstop Filer: The Bandstop filter selectively blocks the passage of frequencies within
certain range, and allows free passage for frequencies outside the range. It is the opposite
of a band-pass filter [1].
 Notch Filter: The notch filter is a bandstop filter with a narrow stop band. The notch
filter is characterized by containing one or more sharp attenuation slope in its frequency
response. This filter type is generally used to remove a disturbance of a known narrow
band frequency. The width of the attenuation notch and the maximum attenuation
depends on the order and implantation of the filter. The maximum gradient of the phase
slope is at the center of the attenuation frequency [1].

Signal Processing Tool

The Signal Processing Toolbox application, SPTool, provides a rich graphical environment for
signal viewing, filter design, and spectral analysis [2]. You can use SPTool to analyze signals,
design filters, analyze filters, filter signals, and analyze signal spectra. You can accomplish these
tasks using four GUIs that you can access from within SPTool:

 The Signal Browser is for analyzing signals. You can also play portions of signals using
your computer's audio hardware (sound card). Using the Signal Browser you can
o View and compare vector/array signals.
o Zoom in on a range of signal data to examine it more closely.
o Measure a variety of characteristics of signal data.
o Play signal data on the audio hardware.

To open/activate the Signal Browser for the SPTool, Click one or more signals (use the Shift
key for multiple selections) in the Signals list of SPTool. Then Click the View button in the
Signals list of SPTool.

 The Filter Designer is for designing or editing FIR and IIR digital filters. Note that the
FDATool is the preferred GUI to use for filter designs [3]. FDATool is discussed later
in this lab.

Page 2 of 26
  The Filter Viewer is for analyzing filter characteristics.
 The Spectrum Viewer is for spectral analysis.

Open SPTool by typing sptool at the command prompt >>.

>>sptool

You see 3 panes - Signals, Filters and Spectra. View all the signals in these panes. You can play
the signals in Signals through using the speaker icon in the Signals browser. (Note: The speaker
will only play the signal between the two cursors on the Signal Browser). When you are trying to
view spectra, note that many methods of determining spectra, including FFT are available. At
this point, just use the FFT. In the Filters pane just View all three filters (LSip, PZip, FIRbp) at
this time and get a sense of the capabilities of Filters.

Filter Design and Implementation

The design of a digital filter is carried out in three steps:

 Specifications: Before we can design a filter, we must have some specifications. These
specifications are determined by the applications.
 Approximations: Once the specifications are defined, we use various concepts and
mathematics to come up with a filter description that approximates the given set of
specifications.
 Implementation: The product of the above step is a filter description in the form of
either a difference equation, or a system function 𝐻(𝑧), or an impulse response ℎ(𝑛).
From this description we implement the filter in hardware or through software on a
computer.

The goal of filtering is to perform frequency-dependent alteration of a signal. A simple design

specification for a filter might be to remove noise above a certain cutoff frequency. A complete
specification determines the amount of passband ripple (𝑅𝑝 , in decibels), stopband attenuation
(𝑅𝑠 , in decibels), or transition width (𝑊𝑠 − 𝑊𝑝 , in hertz).

Page 3 of 26
Figure 1. Specifications for a realizable filter

Analog Filter Design Using MATLAB

MATLAB provides many filter design tools [3][4]. Most of the tools are aimed at digital filter
design, but some of the tools also support analog filter design. We briefly consider some analog
filter design tools. Here we briefly summarize the characteristics of the lowpass version the most
widely used analog filters in practice: Butterworth, Chebyshev (Type-1), and Elliptic.

 Butterworth Filter: This filter is characterized by the property that its magnitude
response is flat in both passband and stopband. The magnitude squared response of N-th
order lowpass filter is given by

1
|𝐻𝑎 (𝑗𝛺|2 =
𝛺 2𝑁
1+( )
𝛺𝑐

where 𝑁 is the order of filter and Ω𝑐 is the cutoff frequency in rad/sec. To design an analog
Butterworth filter using MATLAB, one uses the command

[b, a] = butter (N, cutoff_freq,’s’)

This command tells MATLAB to design a Butterworth filter of order 𝑁 and cutoff frequency
cutoff_freq. The 's' tells MATLAB to design an analog filter. (Without this command,

Page 4 of 26
MATLAB designs a digital filter.) The vectors a and b hold the coefficients of the denominator
and the numerator (respectively) of the filter's transfer function.

Example 1. Giving MATLAB the commands

>> [b, a] = butter (4, 100,’s’);

>> g = tf(b, a)

causes MATLAB to respond with

Giving MATLAB the command

>> bode(g,{30,4000}); grid;

causes MATLAB to respond with Figure 2. (The term {30,4000} in the bode command causes
the command to plot the magnitude and phase response from 30 rad/s out to 4,000 rad/s.) . Two
points are worth noting. Looking at the magnitude plot, one sees that at 100 rad/s the response
seems to have decreased by about 3 dB- as it should. Also, one notes that from 100 rad/s to 1000
rad/s the response seems to drop by about 80 dB. As this is a fourth order filter its should be 𝟒 ×
𝟐𝟎 dB/dec.

Page 5 of 26
Figure 2. The magnitude and phase plot of the fourth-order low-pass Butterworth filter with a
cutoff frequency of 100 rad/s.

 Chebyshev Filter: There are two types of Chebyshev filters. The Chebyshev-I filters
have equiripple response in the passband, while the Chebyshev-II filters have equiripple
response in the stopband. The magnitude-squared response of a Chebyshev-I filter is
1
|𝐻𝑎 (𝑗𝛺|2 =
𝛺 2𝑁
1 + 𝜀 2 𝑇𝑁2 (𝛺 )
𝑐

where 𝑁 is the order of the filter, 𝜀 is the passband ripple factor, which is related to 𝑅𝑝 , and
𝑇𝑁 (𝑥) is the Nth-order Chebyshev polynomial given by

cos(𝑁 cos −1 (𝑥)) , 0≤𝑥<1 Ω

𝑇𝑁 (𝑥) = { 𝑤ℎ𝑒𝑟𝑒 𝑥=
cos(cosh−1 (𝑥)) , 1<𝑥<∞ Ω𝑐

MATLAB provides functions called cheby1 and cheby2 to design Chebyshev- type I and
Chebyshev- type II respectively. Read more about these functions (doc cheby1, doc
cheby2).

 Elliptic Filter: These filters exhibit equiripple behavior in the passband as well as in the
stopband. They are similar in magnitude response characteristics to the FIR equiripple
filters. Therefore elliptic filters are optimum filters in that they achieve the minimum

Page 6 of 26
 order 𝑁 for the given specifications. The magnitude-squared response of an elliptic filter
is given by
1
|𝐻𝑎 (𝑗𝛺|2 =
𝛺 2𝑁
1 + 𝜀 2 𝑈𝑁2 (𝛺 )
𝑐

where N is the order of the filter, 𝜀 is the passband ripple factor, which is related to 𝑅𝑝 , and
𝑈𝑁 (. ) is the Nth-order Jacobian elliptic function. MATLAB provides a function called ellip to
design Elliptic filter.

Finite impulse response (FIR) and Infinite impulse response (IIR) Filters in
MATLAB
In practice we would prefer either a rational system function corresponding to FIR or IIR filters.
Thus we consider using the difference equation model given in (1) and (2) representing the
descriptions in the time- and frequency-domain, respectively.

𝑁 𝑀

𝑦(𝑛) = − ∑ 𝑎𝑘 𝑦(𝑛 − 𝑘) + ∑ 𝑏𝑘 𝑥(𝑛 − 𝑘)

𝑘=1 𝑘=0 (1)

∑𝑀
𝑘=0 𝑏𝑘 𝑧
−𝑘
𝐻(𝑧) =
1 + ∑𝑁
𝑘=1 𝑎𝑘 𝑧
−𝑘
(2)

In general, IIR filters are less complex than FIR filters (the difference being that for FIR filters
there is a restriction that 𝑎𝑘 = 0 all 𝑘), as they require fewer parameters and less memory for the
same “quality” of filter performance. IIR filter design methods differ from FIR primarily in how
performance is specified. For loosely specified requirements, a Butterworth filter is often
sufficient. More rigorous filter requirements can be met with the Chebyshev and elliptic filters.

In addition, the primary advantage of IIR filters over FIR filters is that they typically meet a
given set of specifications with a much lower filter order than a corresponding FIR filter. This
has the obvious implementation advantages. IIR filters have nonlinear phase. However, if data
processing is to be performed offline, then the effects of the nonlinear phase can be eliminated.
So let us assume that the entire input data sequence is available prior to filtering. This allows for

Page 7 of 26
a non-causal, zero-phase filtering approach (via the filtfilt function), which eliminates the
nonlinear phase distortion of an IIR filter. (doc filtfilt)

In practice one wonders about which filter (FIR or IIR) should be chosen for a given application
and which method should be used to design it. Because these design techniques involve different
methodologies, it is difficult to compare them. One basis of comparison is the number of
multiplications required to compute one output sample in the standard realization of these filters.

The Signal Processing Toolbox used functions to estimate the minimum filter order that meets a
given set of filter specifications. Luckily, MATLAB has graphical user interface (GUI) filter
design program, which requires us to fill a few fields and to click a few buttons.

Filter Configurations
First, recall that in digital signal processing (DSP) we are dealing with sampled signals, so we
have to normalize the frequencies to the Nyquist frequency (defined as one-half the sampling
frequency). All the filter design functions in the Signal Processing Toolbox operate with
normalized frequencies, so that they do not require the system sampling rate as an extra input
argument. The normalized frequency is always in the interval 0 ≤ 𝑓 ≤ 1. For example, with a
2000Hz sampling frequency, 400 Hz is 400/1000 = 0.4. To convert normalized frequency to
angular frequency around the unit circle, multiply by 𝜋. To convert normalized frequency back
to Hertz, multiply by half the sample frequency. So we have the following:

 Lowpass filters remove high frequencies (near 1).

 Highpass filters remove low frequencies (near 0).
 Bandpass filters pass a specified range of frequencies between 0 & 1
 Bandstop filters remove a specified range of frequencies between 0 & 1

Filter Specifications in MATLAB

 𝜔𝑝 - Passband cutoff frequencies (normalized)

 𝜔𝑠 - Stopband cutoff frequencies (normalized)
 𝑅𝑝 - Passband ripple: deviation from maximum gain (dB) in the passband
 𝑅𝑠 - Stopband attenuation: deviation from 0 gain (dB) in the stopband

Page 8 of 26
Recall the concept of the ideal lowpass filter, which is simply visualized in the frequency domain
as the rectangle function (of a specified width and centered at the origin). Also recall that this
filter, although handy in theory, is not implemented in practice. There are several reasons why
ideal (lowpass, bandpass, highpass, and band-stop) filters are not used in real-life:

1. The impulse response, ℎ(𝑛), is non-causal as a consequence of the Paley-Wiener

theorem, which implies these filters cannot be implemented in practice on a DSP.
Another way to think about this is that ℎ(𝑛) has infinite support for any of the ideal
filters. Thus, it cannot be zero for 𝑛 < 0 making it necessarily non-causal.
2. When implemented in software or hardware, due to the finite number of elements
employed for processing, an ideal filter exhibits the undesirable Gibbs phenomenon. In
signal and image processing, this shows up as the infamous ringing effect.

To bypass the above inadequacies, the design of digital filters using the windowing technique is
an alternative to ideal filters. These filters can be implemented as FIR filters, and make use of the
well-known Bartlett, Blackman, Hamming, Hanning, and Kaiser windows. The shortcoming of
digital filter design via windowing techniques is that one cannot individually control the design
parameters of the filter. For example, in Figure 1, 𝜔𝑝 (passband frequency) and 𝜔𝑠 (stopband
frequency) cannot be independently controlled using the windowing technique. By making the
transition band (frequencies between 𝜔𝑝 and 𝜔𝑠 ) narrower, you must make a sacrifice in the
form of larger undesirable ripples in the passband (frequencies between 0 and 𝜔𝑝 ) and stopband
(frequencies exceeding 𝜔𝑠 ); that is smaller 𝜔𝑠 − 𝜔𝑝 necessitates 𝛿1 and 𝛿2 larger. These tradeoffs
are all due to bypassing the two inadequacies of ideal filters.

Filter Design and Analysis Tool

In this section, you will learn how to use MATLAB’s handy filter design and analysis tool
(FDAT) [3]. To begin, start MATLAB. Now, enter fdatool into the command window. This
should, when executed, bring up the FDAT’s GUI, shown in Figure 3 below. The process of
designing a filter is fairly self-explanatory: you simply set all of the filter specifications in the
lower half of the GUI. When you are satisfied with your specifications, click on the Design Filter
button. The magnitude response of the resulting filter will appear in the Magnitude Response
pane. Note that you can view the coefficients of the filter’s transfer function in second order

Page 9 of 26
sections by clicking on the Filter coefficients button at the top of the GUI (which looks like
[b,a]).

Figure 3. Filter Design and Analysis Tool

One useful feature of FDAT is that you can store multiple filters at once and switch between
them as you wish. After you have designed a filter, you can store it by selecting the Store Filter
button; this will prompt you to enter a name for the filter. Once you have stored the filter, you
can begin designing a new filter by choosing new filter specifications. To access previously
stored filters, click on the Filter Manager button. Finally, you can save all of your stored filters
in one “session” by selecting FileSave Session As; this will save your session with a .fda
extension. You can always open previously saved sessions in the FDAT GUI. Another feature of
FDAT that we will be using is to export filters to a Simulink model as a single-input, single-
output block. To do this, make sure the filter you want is currently shown in the GUI (if not,
switch to it using the Filter Manager). Then, click on FileExport to Simulink Model; a new
set of options should appear in the lower half of the GUI. Give the filter a good, descriptive
Block Name and make sure the Destination is Current. Press the Realize Model Button (this
only works if you have a Simulink Model currently open, of course).

Page 10 of 26
Simulink
Simulink is an environment for simulation and model-based design for dynamic and embedded
systems. It provides an interactive graphical environment and a customizable set of block
libraries that let you design, simulate, implement, and test a variety of time-varying systems,
including communications, controls, signal processing, video processing, and image processing
[5]. In addition, you do not specify exactly how signals are computed in Simulink. You simply
connect together blocks that represent systems. These blocks declare a relationship between the
input signal and the output signal. The Simulink excels at modeling continuous-time systems. Of
course, continuous-time systems are not directly realizable on a computer, so Simulink must
simulate the system. The Simulink can also model discrete-time systems, and mixed discrete and
continuous-time systems.

To open the Simulink library browser, first you need the MATLAB to be running. Start
MATLAB, and then in the MATLAB Command Window, enter simulink. Then, the Simulink
library browser opens (Figure 4). You can also open the Simulink Library Browser by clicking
the Simulink icon on the MATLAB toolbar.

Figure 4. Simulink library browser

Page 11 of 26
The Simulink library browser displays the block libraries installed on your computer. You can
start building models by copying blocks from a library into a Simulink editor window. To create
a new Simulink model;

1) From the Simulink library browser menu, select FileNew  Model. An empty model
opens in the Simulink editor window.
2) Select FileSave, in the Save As dialog box, enter a name for your model, and then
click Save. Simulink saves your model with the file extension .mdl or .slx.

The Simulink editor window contains a block diagram of your model. You can build models in
the editor by dragging blocks from the Simulink Library Browser window, arranging the blocks
logically, connecting the blocks with signal lines, and then setting the parameters for each block.
The Simulink editor window also allows you to:

 Set configuration parameters for the model, including the start and stop time, type of
solver to use, and data import/export settings.
 Start and stop a simulation of the model.
 Save the model.
 Print the block diagram.

Before you simulate a model, you have to set simulation options. Specify options using the
Configuration Parameters dialog box. In the Simulink editor window, select Simulation 
Configuration Parameters. The Configuration Parameters dialog box opens to the Solver pane.
After entering your configuration parameters, you are ready to simulate the simple model and
visualize the simulation results. In the Simulink editor window, select Simulation  Start from
the menu. The simulation runs, and then stops when it reaches the stop time specified in the
Configuration Parameters dialog box. Alternatively, you can control a simulation by clicking the
Start simulation button and Stop simulation button on the editor window toolbar.

Page 12 of 26
Importing Audio into MATLAB Workspace

Representing, Playing and Plotting Sampled Sound Signals in MATLAB

In MATLAB, mono sounds can be represented as a really long vector, and stereo sounds as two
really long vectors put together. You can edit and modify the vector in order to change the
sound. It should be noted that the short sound signal in MATLAB requires a very long vector.
For example, suppose we wanted to represent a 3-second recording of a sound by sampling the
sound (recording a value) once every 2 ms. This means we are recording 500 numbers every
second, so if we did it for 3 seconds we would need a vector of length 1500 to represent all the
numbers in MATLAB. Fortunately, MATLAB can handle fairly big vectors. MATLAB has
some sound signals already built-in. You will need your headphones for this part. To load and
play one of these sounds, type

>> load handel

>> sound(y,Fs)

The command loads a variable 𝑦, which contains a vector of 73113 elements. This song is about
9 seconds long, with 8192 samples each second (the sampling rate is stored in the variable 𝐹𝑠 ).
Recall that you can specify a part of a vector using a range of indices with the colon operator.
Play the first 1/3 of the recording by typing:

>> L=length(y)
>> sound(y(1:L/3),Fs)

Sounds can be on your computer in different formats. For example, .wav and .mp3 files are two
particular formats for storing sounds, and sound-playing programs know how to read the files
and produce sound using the computer’s sound device. These formats all store a sampled signal,
so the song is really just a long list of numbers, i.e. values of the signal at each sample time.
There are a number of ways that we can get sounds into MATLAB, including:

 Convert an external sound file into a MATLAB vector, e.g. using the wavread
command for sound files stored in the .wav format, as in either:

>> [mySound Fs] = wavread(‘somesound.wav’);

>> [mySound Fs] = wavread(‘somesound’);

Page 13 of 26
  Load a sound signal that already exists as a MATLAB vector into your workspace, using
the load command. The syntax is either:

>>load handel;
>> load(‘handel’);
(The signal and sampling frequency are put into previously-defined variables, in this case 𝑦 and
𝐹𝑠 . The usual value of 𝐹𝑠 for built-in MATLAB sounds is 8192 Hz.)

 Create a vector from scratch in MATLAB.

 Use the wavrecord function in MATLAB to record sound for the audio input of your
sound card.

We will use one of these methods to generate vectors which store sounds. Once you have
generated a vector, the original format doesn’t matter, so plotting and playing sounds is the same
for all cases. Format will matter again when you want to “write out” the sound (save it in a file
for future use). You can play a vector as a sound using the sound command. This command
requires values to be in the range [−1, 1] or it will clip them to this range. (You can use
soundsc instead to automatically scale values to this range. This avoids clipping, but changes
the loudness of the sound.) If you want to play a sampled sound (in MATLAB or with other
tools), you need to specify the playback rate (sampling rate) 𝐹𝑠 in samples per second (Hz).
(Recall 𝐹𝑠 = 1/𝑇𝑠 , where 𝑇𝑠 is the time between samples). In MATLAB, the function sound
allows you to specify the sampling frequency as the second argument:

>> sound(mySound,Fs);
If you don’t specify anything, it will use the default sampling frequency of 8192Hz (which was
just fine for the Handel example). To convince yourself that this is important, see the prelab
question1. You can plot the time signals using either plot or stem, where plot hides the
discrete nature and stem makes it explicit. When you are plotting sampled signals as time
functions, you should make sure that the appropriate time information is displayed on the time
axis, for example using:

>> t=0:Ts:2;
>> plot(t,y);
where 𝑇𝑠 = 1/𝐹𝑠 . Another thing to keep in mind is that a very long signal will be hard to view in
a single plot, especially a sinusoid with constant amplitude. It is often useful to plot only portions

Page 14 of 26
of a signal, or plot the signal in sections. You can display multiple plots at once using the
function subplot(n,m,k), which creates a 𝑛 × 𝑚 matrix of plot spaces in the figure, you
can save it in the .wav format by using:

>> wavwrite(mySound,Fs,‘filename’);
where mySound is the vector your sound is stored in, 𝐹𝑠 is the appropriate sample rate, and
“filename.wav” is the name you want the file to have. If you have values outside [−1,1],
they will be clipped and the function will return a warning message. If you don’t specify the
sampling frequency, it will assume a default of 8192Hz. You can also specify the number of
bits/sample, but we will just stick with the default (16 bits). If you want to write a stereo file, the
format should be a matrix with 2 columns, one per channel.

Example 2. Reading stereo sound:

[audio,fs]=wavread('name_of_wav_file.wav'); % transfers .wav

file into array audio and reading the sampling frequency to fs
x = audio ( : , 1 ); % assign left channel data to x
y = audio ( : , 2 ); % assign right channel data to y
Ts=(1/fs)*length(x); % Calculates the duration of .wav file
t=linspace(0,Ts,length(x)); % Interpolates ‘t’ from 0 to ‘time’
so creates a time vector
sound(audio,fs) % Plays the array audio with a sampling rate fs

Since sound signals are represented as vectors in MATLAB, you can do any mathematical
operation on the sound signals that you could do on elements in a vector. In other words, you can
create your own sounds with MATLAB scripts and functions. You will get to make a sound
composition by modifying, mixing and stringing sounds together. (doc soundmixer).

Importing a Multimedia File to Simulink

In Simulink library, there is a specific block to read multimedia files named as From Multimedia
File. Any sound or image data supported by MATLAB can be imported into Simulink model by
using this block. Loaded audio file can be listened with the block named To Audio Device. If the
loaded multimedia file is image or video, the Video Viewer block should be used.

Page 15 of 26
Figure 5. Import audio file to Simulink model

Prelab
Question 1. The MATLAB function sound (see help sound) with syntax

>> sound(sampledSignal, frequency)

sends the one-dimensional array or vector sampledSignal to the audio card in your
computer. The second argument specifies the sampling frequency in Hertz. The values in
sampledSignal are assumed to be real numbers in the range [−1, 1]. Values outside this
range are clipped to −1 or 1.

Download the audio file “lab_5_Audio_1.wav” from the course directory /groups/e/elec364_1
(see Laboratory Guidelines).

Use wavread command to read the audio file “lab_5_Audio_1.wav”

>> [y,fs]=wavread('lab_5_Audio_1.wav');

After reading, listen to

a)

>> sound(y,fs)
b)

>> sound(0.25*y,fs)

Page 16 of 26
and

>> sound(4*y,fs)
Explain in what way these are different from what you heard in the part (a).

c)

>> sound(y, fs/2)

and

>> sound(y, fs*2)

Explain how these are different from what you heard in the part (a).

Question 2. Designing the Filters Using FDATOOL, perform the following tasks, assuming
a sampling rate of 8 kHz:

I. Design a minimum order, stable, lowpass Butterworth filter with a passband frequency of
1 kHz and a stopband frequency of 1.4 kHz. Make the attenuation 1 dB at the passband
frequency and 80 dB at the stopband frequency.
II. Design a minimum order, stable, lowpass Chebyshev Type I filter with the same
specifications as the Butterworth filter.
III. Design a lowpass FIR filter using the Blackman Window with a cutoff frequency of 1
kHz. Specify the order of the filter such that the first minimum in the stopband (preceding
the first lobe) is as close to 1.4 kHz as possible without exceeding it.

Now answer the following

a) What is the order of the lowpass Butterworth filter you designed?

b) What is the order of the lowpass Chebyshev Type I filter you designed?
c) Compare the implementation cast of each of the 3 filters. Do you see how inefficient the
windowing technique is? How much more expensive in terms of memory is the
windowing technique from the best IIR filter?
d) For the filter you designed in Part I, round the filter coefficients (b,a) to the nearest
integer value. Use the MATLAB round command:

>> help round

Page 17 of 26
round rounds towards the nearest decimal or integer

round(X) rounds each element of X to the nearest integer.

round(X, N), for positive integers N, rounds to N digits to

the right of the decimal point. If N is zero, X is rounded to
the nearest integer. If N is less than zero, X is rounded to
the left of the decimal point. N must be a scalar integer.

Obtain the pole-zero plot of the filter with the original coefficients (use the zplane(b,a)
command where b and a are the filter coefficients) and the pole-zero plot of the filter with the
rounded values of the coefficients. Comment on any differences between the two plots.

In this exercise, we intentionally rounded the coefficients using the MATLAB round
command. As alluded to in Example 4 of Lab 2, (unintentional) rounding may result from
the way real numbers are represented inside a computer. Here is Example 4 from Lab 2:

clear
x = 0.1 + 0.1 + 0.1
if ( x == 0.3)
disp(‘ x is equal to 0.3’)
else
disp(‘x is is NOT equal to 0.3’)
end
% use the fprintf output command to display the value of x to
% 18 decimal place in a field-width of 21 columns
fprintf(‘ x is really = %21.18f\n’, x)

Run the above MATLAB code and see the effects of rounding for yourself. Comment on the
source of the rounding error when a real number is stored inside a computer. HINT: IEEE
754 floating point representation. ANOTHER HINT : 1/3 = 0.33333333333333333333…

Page 18 of 26
 Question 3. Creating a Noisy Signal using Simulink

You will create a sinusoid single corrupted by high frequency noise. Both the sine wave and the
noise will be discrete; thus, your whole Simulink model will be discrete (this is purposely done
to simplify things). Here’s how to create such a signal:

I. Add a Sine Wave block to your model. Set the frequency to 100 Hz and the Sample time
to 1/8000.
II. Add a Uniform Random Number block to your model. Set Sample time to 1/8000 and
Maximum and Minimum to 5 and -5, respectively. As its name suggests, this block
outputs random numbers, which for our purposes is a good model for noise.
III. We want to restrict our noise to only have high frequency components. To do this, we are
going to send it through a highpass filter. Add a Highpass Filter block and open the Block
Parameters. Under Filter specifications, set Impulse response to IIR and Order mode to
Minimum. Under Frequency specifications, set Frequency units to Hz, Input Fs to 8000,
Fstop to 3600, and Fpass to 3800. Under Magnitude specifications, set Magnitude units to
dB, Astop to 60, and Apass to 1. Finally, under Algorithm, set Design method to Elliptic
and Structure to Direct-form II SOS. Right now, all that you have to understand about
this filter is that it only passes frequencies above 3800 Hz (Fpass); anything below 3600
Hz (Fstop) is attenuated by at least 60 dB (the region between these two frequencies is a
transition region).
IV. Send the random number signal from Step II through the highpass filter and add it to the
sine wave you made in Step I.
V. To view the noisy sine wave signal, use Scope blocks. Also, use another scope to view
the original sine wave. Use the Scope parameters to set ‘limit data points to last’
parameter and to save the data to your workspace.

Page 19 of 26
Useful MATLAB Commands
You can use MATLAB sim command to Run Simulink model from the script file.

Tools: Sumlink, sptool, fdatool

Sound: wavread, wavwrite, sound, soundsc, soundmixer, audioplayer
Filter: butter, buttap,cheby1, cheb1ap, cheby2, cheb2ap, ellip,
ellipap, besselap
Frequency Domain: fft
Signal manipulation: filter, filtfilt
Elementary Matrices and Matrix Manipulation: i, ones, pi, rand, randn,
zeros.
Two-Dimensional Graphics: axis, grid, legend, plot, stem, title,
xlabel, ylabel.
Elementary Functions: cos, sin, exp, imag, real,abs.
Data Analysis: sum

Page 20 of 26
Lab problems
During the lab, you must show your results/outputs of each problem to your Lab
demonstrator/TA. After the lab, each student must submit a lab report contains results,
discussion, MATLAB codes and Simulink models. Also, at the end of the lab report attach the
results and solutions to the prelab questions.

Problem 1. Consider the input signal

𝑥(𝑡) = sin(100𝑡) + sin(2000𝑡) + sin(6000𝑡)

We need to pass the term sin(2000𝑡) and to attenuate the other terms.

a) Design a digital fourth and eighth-order IIR Butterworth filters. Plot the input signal
along with the magnitude response of the filter and its output signal for each filter.
b) Design a digital fourth and eighth-order IIR Chebyshev type I filters. Plot the input signal
along with the magnitude response of the filter and its output signal for each filter.
(consider 𝑅𝑝 = 1 𝑑𝐵).
c) Design a digital fourth and eighth-order IIR Elliptic filters. Plot the input signal along
with the magnitude response of the filter and its output signal for each filter. ( consider
𝑅𝑝 = 1 𝑑𝐵, 𝑅𝑠 = 60 𝑑𝐵)
d) For parts a), b), and c) plot the discrete Fourier transform of the input signal and the
output signal for eighth order filters, you can use FFT command.

Hint:

 You have to choose a sampling frequency.

 When you use the MATLAB functions butter, ellip, cheby1, we need to
normalize this digital passband to lie in the interval [0 1].

Page 21 of 26
 Problem 2.
Part A: Create a signal with three seconds duration. The signal contains three tones of equal
amplitude, as following
a) Contain a 200 Hz tone for 0 < 𝑡 < 3.
b) Contain a 330 Hz tone for 1 < 𝑡 < 3.
c) Contain a 480 Hz tone for 2 < 𝑡 < 3.

Note: The signal should have a sampling frequency equal to 8192 Hz.

View the signal before proceeding (in time domain and in frequency domain), and listen to the
signal using soundsc command.

Part B: Design a filter to remove the 330 Hz component from the main signal.

a) Plot the designed response (magnitude and phase) of the filter.

b) Filter the main signal with the designed filter.
c) Compare signals and spectra (in time domain and in frequency domain) before and after
filtering, and listen to the filtered signal using soundsc command.

Problem 3. Building Simulink model to read an audio file and manipulate the sound by
adding upsamping or downsampling block.
a) Download the audio file “ELEC_364_lab_5_Audio_S.wav” from course directory
/groups/e/elec364_1 (see Laboratory Guidelines).
b) Open new Simulink model window, and import the audio file
“ELEC_364_lab_5_Audio_S.wav”. Set Audio output data type to double.
c) Add To Audio Device block and connect the output of the audio block to the input of the
To Audio Device block. Click run button and listen.

Page 22 of 26
Part A:

d) After listening, add Downsample block, and set downsample factor k to 8.

e) Build the Simulink model shown in Figure 6. Set the Scope parameter “limit data point to
last” to 327680.

Figure 6. Problem 3-Part A, Simulink Model

f) Click run button, and listen to the sound. Did you notice any difference; write that in your
lab report. Use single figure to plot the original sound signal in time and frequency
domain (use FFT) and audio_down_sampled signal in time and frequency domain (use
FFT), you output results should be similar to Figure 7.
g) Now, reset the downsample factor k to 16 and repeat the step f).

Page 23 of 26
Figure 7. Problem 3-part A, plot of the original audio signal and the down sampled audio signal
in time and frequency domain.

Page 24 of 26
Part B:

h) Build the Simulink model shown in Figure 8, and add Upsample block. Set the upsample
factor L to 8, and set the Scope parameter “limit data point to last” to 327680.

Figure 8. Problem 3-Part B, Simulink model

i) Double click of the To Audio Device block, and uncheck the Inherit sample rate from
input option.
j) Click run button, and listen to the sound. Did you notice any difference; write that in your
lab report. Use single figure to plot the original sound signal in time and frequency
domain (use FFT) and audio_up_sampled signal in time and frequency domain (use
FFT), you output results should be similar to Figure 9.
k) Set the upsample factor L to 16 and repeat the step j).

Page 25 of 26
Figure 9. Problem 3-part A, plot of the original audio signal and the up sampled audio signal in
time and frequency domain.

References
[1] Signals and Systems, 2nd ed. , A.V. Oppenheim and A.S Willsky, Prentice-Hall, ISBN 0-13-
814757-4, 1997.
[2] http://www.mathworks.com/help/signal/ug/sptool-an-interactive-signal-processing-
environment.html.
[3] http://www.mathworks.com/products/signal/code-
examples.html?file=/products/demos/shipping/signal/introfdatooldemo.html.
[4] http://www.mathworks.com/discovery/filter-design.html.
[5] http://www.mathworks.com/academia/student_center/tutorials/simulink-launchpad.html.

Page 26 of 26