The first seven weeks of the course will be structured labs based on fundamental digital signal processing (DSP) concepts from ECE 310 . The next two weeks will be on the implementation and simulation of a fundamental DSP algorithm of a student's choosing from a set of seminal DSP papers (such as adaptive filtering, pitch detection, edge-aware filtering, motion tracking, pattern recognition, etc). The remaining six weeks in the course will revolve around the development, testing, and documentation of a DSP project of the student's choice (subject to instructor approval).
Students will learn to prototype, implement, and analyze real-time mobile DSP systems. Students will both broaden and deepen their understanding of basic DSP theory and techniques and learn to relate this understanding to real-world observations and applications. Students will learn industrially-relevant skills such as rapid design prototyping in Python, and Android development of DSP applications in C++/Java for computationally-constrained mobile devices. Other significant educational experiences include open-ended design, oral, and written communication, and team projects.
Lend-lease Tablet: Please read the following link carefully and request a tablet per student. An Android device is required for this course. If you are willing to use your own Android device, then you do not need to request. Please keep all packaging for the return shipment! https://ece.illinois.edu/academics/ugrad/lab-kits
Location: ECEB 3013
Time: Monday, 1:00-1:50 PM
Location: ECEB 5072
Time
ABA: Tuesday, 2:00-3:50 PM
ABC: Wednesday, 2:00-3:50 PM
ABD: Thursday, 2:00-3:50 PM
ABE: Friday, 2:00-3:50 PM
| Week of | Topic | Lab | Due(in lab) |
|---|---|---|---|
| 1/20 | No Lecture (MLK) | N/A | N/A |
| 1/27 | Lec 1 - Course Overview, slide | Lab 1 - Build your first Android App | Mock Quiz (extra credit) |
| 2/3 | Lec 2 - Audio Processing | Lab 2 - Real-time Audio Filtering | Prelab 2, Demo(lab1), Quiz 1 |
| 2/10 | Lec 3 - Short-time Spectral Analysis | Lab 3 - Spectrogram | Prelab 3, Demo(lab2), Quiz 2 |
| 2/17 | Lec 4 -Correlation Analysis | Lab 4 - Pitch Detection | Prelab 4, Demo(lab3), Quiz 3 |
| 2/24 | Lec 5 -Pitch Modification | Lab 5 - Pitch Synthesis | Prelab 5, Demo(lab4), Quiz 4, Early prototype proposal due 2/24 |
| 3/3 | Lec 6 -Overview of 2D Image Processing | Lab 6 - Image Processing | Prelab 6, Demo(lab5), Quiz 5 |
| 3/10 | Lec 7 -Video tracking, KCF | Lab 7 - Video Processing | Prelab 7, Demo(lab6), Quiz 6 |
| 3/17 | No Lecture (Spring break) | N/A | N/A |
| 3/24 | Lec 8 -Handwritten digit recognition: Part 1 | Prototype Latex Template | Demo(lab7), Quiz 7, Prototype proposal due 3/24 |
| 3/31 | Lec 9 -Handwritten digit recognition: Part 2 | Prototype | N/A |
| 4/7 | Demo week | Prototype demo and presentation | Presentation File, prototype code, Final proposal due |
| 4/14 | TBD | Final Project | N/A |
| 4/21 | TBD | Final Project | Milestone 1 |
| 4/28 | TBD | Final Project | Milestone 2 |
| 5/5 | Demo week | Final demo and presentation | Final Project Report, Presentation File, Source Code due |
There are seven 15-minute quizzes (plus one mock quiz) throughout the semester. They are open-book individual assessments taken at Prairielearn . Each quiz starts at the beginning of each lab section and ends after 15 minutes (e.g. AB1 quiz opens 2:30-2:45, every Tuesday). Students MUST take the quiz at their registered lab section. There is no makeup for missed quizzes. An absence letter from the Dean of Students is required to waive a missed quiz due to acute medical condition. Discussion of the quiz is NOT allowed until all sections have completed the quiz. The grading will be published every Friday evening. You will earn extra credit for the mock quiz (make-up for lost points in the quizzes).
Students will be working in groups to complete all labs and final project. Typically, groups of two are strongly preferred, group of more or less is allowed only on rare occasions.
For structured labs (lab 1 ~ lab 7), groups will be formed randomly and differently for each lab, so that students could have the chance to work with different partners.
For assigned project labs and final project labs, students are expected to form their own groups. Feel free to form groups across different sections; if you plan to do so, make sure the entire group can attend one of the sections because you will need to do presentations and demos as a whole.
All pre-lab, lab work, and project-related documents must be submitted via Gradescope. Grades will also be published on Gradescope once they are ready.
Gradescope Entry code: J7PDY7
Campuswire Entry code: 5796
G = Group work, I = Individual work
Structured Labs: 40%
Prelabs (I): 10%
Labs (G): 20%
Lab quizzes (I): 10%
Extra-credit lab 8 (I): 3% = prelab 1.5% + lab 1.5%
Prototype: 20%
Early prototype proposal (I): 2%
Prototype proposal (G): 8%
Prototype demo (G): 5%
Prototype presentation (G): 5%
Final project: 35%
Final project proposal (G): 8%
Milestone 1 & 2 (G): 2.5% + 2.5%
Final demo (G): 7%
Final presentation (G): 8%
Final report (G): 7%
Lecture participation (I): 5%
The structured laboratory segment will count for 40% (10% for prelab, 10% for quiz, and 20% for lab) of the total grade, based on completion of, and oral examination over, the weekly laboratory assignments, including the underlying theory, details of the implementation and code, and the observed behavior of the system. We emphasize that your grade is based heavily on your understanding and demonstration of the course material, not just on submitting working code.
The prototype (based on the student's chosen DSP paper) will account for 20% of the total grade, with 3 % on the early prototype proposal, 7% on the prototype proposal, 5% on the demo, and 5% on the oral presentation.
The final project will count for 35% of the total grade, with 7% on the project proposal, 5% for demonstrations of 2 project milestones, 7% for the final demo, 8% for the oral presentation, and 8% on the final report.
The final 5% of the total course grade comes from lecture participation.
It is expected that each student will attend and participate in scheduled class and laboratory meetings, or will make prior alternate arrangements with the instructor. The final grade may be penalized if this does not occur.
A late penalty of 50% will be assessed for assignments less than a week late; assignments more than a week late will receive no credit. However, all graded assignments must be submitted to receive a passing grade in the course.
Refer to the Early prototype proposal, Prototype, and Final Project pages.
Prof. Thomas Moon: After lecture or by appointment
TA Dimitrios Gotsis (Thu/Fri section): 1-2 pm Thursday
TA Jorge Chavez (Tue/Wed section): 10-11 am Wednesday
Prof. Thomas Moon: tmoon@
TA Dimitrios Gotsis: gotsis2@
TA Jorge Chavez: jorgejc2@
Printed and online sources are allowed with proper citation. Please direct your question to Google or the course staff before you ask your classmates. Given the range of the material for this course, we allow you to refer to any online source, but do not directly copy and paste .
We do not allow inter-group cooperation for the final project. If there is a sign of cooperation between groups, those groups will be treated as a big group, and the grade will be divided accordingly.
More information: Student Code.
Refer to the more course policy page here.