EECS
644 Introduction to Digital Signal
Processing
Catalog Data: EECS
644 (3) Discrete time signal and systems theory, sampling theory, z-transforms,
digital filter design, discrete Fourier transform, fast Fourier transforms, and
hardware considerations.
Prerequisites: EECS
361 Signal & System Analysis or similar background in Fourier signal
analysis (series and transform); linear system analysis (continuous and
discrete); z-transforms, analog and digital filter analysis; analysis and
design of continuous and discrete time system using MATLAB.
Instructional
Mode:
In-person class, 3150 LEA, TR 8:00 - 9:15
Class
Web Page: http://www.ittc.ku.edu/~sdblunt/644/EECS644.htm
Instructor: Dr. Shannon D. Blunt
3034 Eaton Hall
357
Nichols Hall (864-7326)
e-mail:
sdblunt@ku.edu
Course Notes:
Will be
posted on the course web page prior to the class in which they will be covered.
Office
Hours: TR
9:15 - 10:30 or by appointment.
Text: Digital Signal Processing:
Principles, Algorithms, and Applications (4th edition) by Proakis
and Manolakis
Grading Scale: The
following plus/minus grading will be used for this course. The lower limit on
these ranges may be reduced as a function of the distribution of the final
scores.
90.0–92.9:
A− 93.0–100: A
80.0–82.9:
B− 83.0–86.9: B 87.0–89.9: B+
70.0–72.9: C− 73.0–76.9:
C 77.0–79.9: C+
60.0–62.9: D− 63.0–66.9: D 67.0–69.9: D+
0–59.9: F
The
following percentages will be used to weigh the components of the course.
Exam #1 - 25%
Exam #2 - 25%
Final Exam - 35%
Homework - 15%
Homework:
Homework is intended to illustrate and reinforce concepts covered in
class. Homework assignments are posted on the class website. There will be
roughly 6-7 homework assignments. Collaboration with classmates is permitted; copying
is not. Homework is due by the beginning of the class period in which it is
due. Solutions will be posted to the
class website after homework is graded.
Computer Usage:
Some homework problems will require the use of Matlab.
Exams:
There will be 2 mid-term exams during the semester and a final (all
in-class). Makeup exams will not normally be given.
Make-ups: Make-up exams are given rarely,
and only if:
1. I am informed IN ADVANCE, and
2. I deem the reason to be sufficiently
meritorious (job interviews and pleasure trips are not). If the reason is illness, I REQUIRE
documentation of the illness from a health-care professional.
Course Objectives: The
primary objective of this course is to introduce methods for processing
discrete-time signals. This includes waveforms that originate as
continuous-time signals. Other objectives are:
- to understand the processes of
analog-to-digital and digital-to-analog conversion;
- to analyze and evaluate the
performance of discrete-time linear systems;
- to acquire familiarity with
digital filters in terms of analysis, design and implementation;
- to understand discrete
Fourier analysis (DTFT, DFT, DFS, FFT) as it applies to discrete-time signals;
- to present several
applications of digital signal processing algorithms.
This
course is intended to provide you with the necessary analytical tools for work
in digital signal processing. This is not a computer course, nor is it a
digital design course. This course is aimed at a higher level - we will try to
address the problem of what we can do to process a signal if we have a computer
to help us. Therefore, our primary emphasis will be on algorithms for
processing waveforms. A strong background in linear systems theory (i.e.
Fourier and Laplace transforms, convolution and system impulse response,
transfer functions, and poles/zero behavior) will prove essential.
Useful Reading:
I suggest everyone look through some of the past issues of the Signal
Processing Magazine which can be found on the IEEE Xplore database (available
through the KU library system). This
will give you a better idea of the numerous research areas in signal processing
and possibly help you get started on your own research/career in the field.
Class decorum: The
School of Engineering is a professional school, and the decorum in this class
will reflect that. You are expected to
arrive on time, leave on time, and act professionally in class. This includes
being intellectually and physically involved in the class.
Attendance Policy: Attendance at all class
meetings is expected. Anything presented
in class is considered required material.
Academic success in this class requires: regular class attendance, doing
the homework, and class participation. There is a strong correlation between
attendance and the course grade.
Academic Misconduct: Instances of cheating will be
referral to the Dean. Cheating includes, but is not limited to:
copying another exam, copying of hardcopy or online solutions or previously
worked homework or exam solutions, having another person do your work, use of
“tutoring” websites like chegg.com.
EdTech Services:
The use of EdTech services (e.g. Chegg.com) for posting or downloading
material for the preparation and/or submission of exams, homework, etc.,
constitutes academic misconduct, which is not tolerated in the School of
Engineering at the University of Kansas. It violates Article 3r, Section 6 of
its Rules & Regulations, and may lead to grades of F in the compromised
course, a note on your transcript, dismissal from the School,
or expulsion from the University of Kansas. When a person signs up to use
EdTech services, the “terms of service” that are agreed to do not
protect the person when an academic unit (e.g. KU School of Engineering)
conducts investigations related to academic misconduct (e.g. plagiarism and/or
cheating). These services retain contact and information (e.g. IP address,
email, time of use) of subscribers and users, which is released upon request.
If you are feeling unsure about an assignment, it is important to use the
allowable resources available to you, such as instructor office hours or the
book.
Video
and audio recording of the EECS 644 class lectures is strictly prohibited.
Course
Evaluation: A
course evaluation will be available to students at the end of the semester.
Students are strongly encouraged to participate.
Special Needs: Any student with a disability requiring
special accommodation should contact the Student Access Center at
https://access.ku.edu/ in order to make arrangements. Members of KU sanctioned
organizations (band, athletic teams, etc.) that have special needs should also
contact the instructor as the need arises.
Course Schedule (subject
to change)
No class October 15, 2024 (Fall break)
No class November 28,
2024 (Thanksgiving break)
Topic
1.
Course/topic
introduction ~ 0.5 weeks
2.
Discrete-time signals
& systems review ~ 1.5 weeks
a.
Linear time-invariant (LTI)
systems
b.
Linear
constant-coefficient difference equations (LCCDE)
c.
Frequency-domain
representations
3.
z-transform ~ 1.5 weeks
a.
Derivation
b.
Properties
c.
Inverse z-transform
4.
Correlation ~ 0.5 weeks
a.
Auto-correlation and
cross-correlation
b.
Properties
5.
Sampling of continuous-time
signals ~ 2 weeks
a.
Periodic sampling
b.
Frequency domain
representation of sampling
c.
Reconstruction of
bandlimited signals
d.
Changing the sampling
rate
e.
Digital processing of
analog signals
---------
End for Test 1 ---------
6.
Transform analysis of
LTI systems ~ 1.5 weeks
a.
Frequency response of
LTI systems
b.
System functions for
LCCDE
c.
Frequency response for
rational system functions
d.
Relationship between
magnitude and phase
e.
All-pass systems
f.
Minimum/maximum/mixed-phase
systems
7.
Structures for
discrete-time systems ~ 1 week
a.
Signal flow
representations
b.
Basic structures for finite
impulse response (FIR) and infinite impulse response (IIR) systems
c.
Finite precision
numerical effects
8.
IIR filter design
techniques ~ 1 week
a.
Design of discrete-time
IIR filters from continuous-time IIR filters
b.
Design by approximation
of derivatives
c.
Design by impulse invariance
d.
Design by bilinear
transform
9.
FIR filter design
techniques & beamforming ~ 1 week
a.
Design by windowing
b.
Standard beamforming
& windowing
---------
End for Test 2 ---------
10. Discrete
Fourier transform (DFT) ~ 1.5 weeks
a.
Discrete Fourier series
b.
Sampling the Fourier
transform
c.
Properties of the DFT
d.
Linear convolution
using the DFT
11. Computation
of the DFT ~ 1 week
a.
Goertzel
algorithm
b.
Decimation of the DFT
in time & frequency
12. Fourier
analysis of signals using the DFT ~ 1 week
a.
DFT analysis of short,
long, and continuous signals
b.
Practical considerations
Comprehensive Final
Exam: Wed. Dec. 18, 2024, 7:30 a.m. - 10:00 a.m.
Late
work: Assignments should be submitted on the indicated due
date. I acknowledge that life happens, and sometimes a deadline cannot be met
because of illness, caregiving responsibilities, work demands, mental health
struggles, and emergencies. In these
cases, I request that you contact me via email as soon as possible to arrange
an alternative due date. If I do not receive any communication from you before
the assignment is due, you will receive a 0 for late assignments.
Changes: Changes announced in class and on
the class web page will supersede these written instructions.