Signal Processing - July 2016 - 87

(a)

(b)

Figure 2. (a) A photo of hardware that allows students to easily develop analog circuit-based filters. (b) A photo of hardware that allows students to
experiment with a control system.

some design components. In the touchtone decoding and filters activity, for
example, students designed a filter using
the MF10 (universal monolithic dual
switched-capacitor filter) chip to decode
a touch-tone signal. The introduction of
the activity included a discussion of the
history of the development of the touchtone telephone and related design issues.
To minimize the time devoted to building the detection filters, a circuit board
with an MF10 filter was given to the
students. The circuit board shown in Figure 2 had the necessary clock generator,
audio jacks, operational amplifiers, and
an area to wire circuit components. Students were tasked with using the MF10
data sheet to create the required filter
to detect a particular touch-tone. Based
on their calculations, students then programmed the board by proper selection
of resistors. Students were also asked to
answer several short answer questions in
paragraph format and analytical questions using mathematics and concepts
that were presented in the course. Examples of questions for this activity include
"Suppose the transmitter was nonlinear,
list the first three harmonics of each
frequency and then compare these harmonics with the original frequencies"
and "What would happen to your detector circuit if the transmitter frequencies
varied by !2%?"

After completing eight weeks of
activities, students were asked to choose
an application-oriented project based on
the signals and systems concepts. During
the last three weeks of the semester, the
students implemented the project using
hardware or MATLAB. Some project
examples include active noise cancellation, a digital graphic equalizer, computing the fast Fourier transform (FFT)
using a cell phone, a voice modulator,
and a voice recognition control system.

been reduced, and some of the activities have been extended to two weeks.
One of the more challenging aspects
of working with realistic signals and systems is that most students have no idea
what to expect in terms of results from
their experiments. During one lab session,
a student was trying to measure a speech
signal but had the SEB configured incorrectly so that it was multiplying the speech
signal with a low-duty-cycle pulse signal.
When the instructor asked the student
what was wrong with the experiment, the
student responded, "I have no idea what
Lessons learned from using
a speech signal is supposed to look like,
application-oriented activities
so how can I know what's wrong?" If stuBoth sets of activities, described in the
dents don't have a general sense of what
previous section, were initially develthey should expect
oped in the 2009-
during the experi2010 time frame but
one of the more
ment, there is no way
have evolved significhallenging aspects of
they can debug the
cantly since their
working with realistic
experiment. Conseinception. Many lesquently, many stusons were learned by
signals and systems is
dents end up missing
vetting the activities
that most students have
the point of what they
with the students. The
no idea what to expect
are supposed to learn
SEB has been used
in terms of results from
from the activity. In
during three ten-week
their experiments.
another activity, stuquarters per year over
dents are asked to
the last seven years
measure the harmonic distortion of the
and has gone through three revisions,
SEB in the first part and then reconfigure
and the activities are nothing like their
the SEB to measure the signal-to-noise
original versions. The other activities
ratio. Quite often, they would submit data
using various platforms have evolved
for the signal-to-noise ratio experiments
as well. The number of activities has
IEEE Signal Processing Magazine

July 2016

Table of Contents for the Digital Edition of Signal Processing - July 2016