14:332:447 Digital Signal Processing Design

Course Catalog Description: 

14:332:447 Digital Signal Processing Design (3)
This course emphasizes the application of basic concepts of digital signal processing (sampling, filtering, spectral analysis, decimation and interpolation), to building (within a MATLAB framework) real and useful systems for a wide range of application areas, with primary emphasis on image and speech processing systems.

Pre-Requisite Courses: 

14:332:346

Co-Requisite Courses: 

Knowledge of MATLAB Programming

Pre-Requisite by Topic: 

1. sampling of signals (A-D, D-A, sampling rate, aliasing)
2. linear system theory (convolution, z-transforms)
3. signal spectral analysis (DTFT, DFT, FFT)
4. digital filtering techniques (structures, fast convolution)
5. digital filter design methods (FIR, IIR Designs)
6. signal representations (time/frequency/convolutional/LP)

Textbook & Materials: 

L. R. Rabiner and R. W. Schafer, Theory and Applications of Digital Speech Processing, Prentice-Hall Inc., 2011

References: 

R.G. Gonzalez, R.E. Woods, S.L Eddins, Digital Image Processing using MATLAB, Second Edition, Prentice-Hall Inc., 2009.

Overall Educational Objective: 

1.To introduce students to the software and hardware design principles involved in designing and implementing (in MATLAB code) useful and practical DSP-based systems.
2. To provide students a theoretical and conceptual base for further (more advanced) study of DSP-based systems.
3. To introduce students to real world signals in the domains of speech, image, vision, radar, sensors, etc.

Course Learning Outcomes: 

A student who successfully fulfills the course requirements will have demonstrated:
1. An ability to understand the theory (basic mathematics and signal processing) behind DSP implementations of speech and image processing systems.
2. An ability to implement simple signal processing systems efficiently using MATLAB code.
3. An understanding of the fundamentals of gray scale and color image processing, and an ability to apply this understanding to real world images, e.g., photos.
4. An understanding of the fundamentals of speech processing, including acoustic-phonetic representations of speech production, and ear processing methods for speech perception.
5. An ability to analyze the spectrum of a time-varying signal using frame-based approaches with sliding analysis windows.
6. An ability to understand how to de-convolve signals using cepstral domain signal processing methods.
7. An ability to understand how to model signals based on linear predictive analysis methods.
8. An understanding of the processes and impacts of aliasing in both the frequency domain (undersampling in time) and the time domain (undersampling in frequency).
9. An understanding of how to use MATLAB code efficiently via vector operations on signals.

How Course Outcomes are Assessed: 

  • In Class and Homework assignments (20%)
  • Term Project (30%)
  • Mid-Term Exam (20%)
  • Final Exam (30%)


N = none S = Supportive H = highly related

Outcome

Level

Proficiency assessed by

(a) an ability to apply knowledge of Mathematics, science, and engineering

H

HW Problems and MATLAB assignments

(b) an ability to design and conduct experiments and interpret data

H

Design problems in Class and HW, Term Project

(c) an ability to design a system, component or process to meet desired needs within realistic constraints such as economic, environmental, social, political, ethical, health and safety, manufacturability, and sustainability

H

MATLAB Projects in Image and Speech Processing

(d) an ability to function as part of a multi-disciplinary team

S

Team-Based Term Projects

(e) an ability to identify, formulate, and solve ECE problems

S

MATLAB Assignments and Term Project

(f) an understanding of professional and ethical responsibility

N

Class lectures

(g) an ability to communicate in written and oral form

H

Homework problems, Term Project

(h) the broad education necessary to understand the impact of electrical and computer engineering solutions in a global, economic, environmental, and societal context

S

DSP system design from basic principles of speech and image processing

(i) a recognition of the need for, and an ability to engage in life-long learning

H

Discussion of DSP concepts and their changing impact on system design

(j) a knowledge of contemporary issues

S

Discussion of DSP systems for real-world problems

(k) an ability to use the techniques, skills, and modern engineering tools necessary for electrical and computer engineering practice

H

DSP Design term project and related team research on the project

Basic disciplines in Electrical Engineering

S

DSP algorithm design, DSP algorithm implementations

Depth in Electrical Engineering

S

Ability to use DSP Design principles across a wide variety of signals

Basic disciplines in Computer Engineering

H

MATLAB Design

Depth in Computer Engineering

S

DSP Term Project, from concept to implementation and class presentation

Laboratory equipment and software tools

H

MATLAB, computer work stations

Variety of instruction formats

H

Lectures, Homework Problems, Term Projects, Office hour discussions

Topics Covered week by week: 

Week 1: Introduction to Digital Signal Processing Design
Week 2: Review of DSP Fundamentals, Preliminary Exam
Week 3: MATLAB Coding of Signal Processing Algorithms

Week 4: Introduction to Speech and Image Processing

Week 5: Image Processing - Color Images

Week 6: Color Image Intensity Transforms and Spatial Filtering

Week 7: Color Image Processing in the Frequency Domain

Week 8: Image Restoration Methods

Week 9: Mid-Term Exam, Submission of Term Project Titles

Week 10: Time-Domain Signal Estimation Methods (Short-Time Processing Representations)

Weeks 11: Frequency-Domain Methods (Short-Time Fourier Transform Representations)

Week 12: Convolutional Analysis Methods (Cepstral Representations)

Week 13: Algorithms for Signal Processing (Signal Detection, Statistical Pattern Recognizer, Periodicity Detection, Resonance Frequency Estimation)

Week 14: Principles of Computer Vision (Guest Lecturer)

Week 15: Term Project Presentations (15 minutes per project)

Week 16: Final Examination

Computer Usage: 

Weekly MATLAB Exercises within homework problem sets; Term Project implemented in MATLAB

Laboratory Experiences: 

Homework assignments using MATLAB and term project implemented in MATLAB code.

Design Experiences: 

50% of the design experience comes from homework exercises in the areas of image and speech processing. The remaining 50% is based on a term project. The students, either singly or in groups of 2 or 3 students, go beyond the material covered in class lectures and build an advanced DSP system in any area of DSP of interest to the student or students in the group. For example, one group of 2 students designed a MATLAB simulation of the Adobe Photoshop Lite software based strictly on signal processing methods learned in this class and the group was able to obtain image processing results comparable to those of the professional software product from Adobe. In addition, the students designed a light weight and easy-to-use Graphical User Interface (GUI) for using the Adobe emulation software, thereby broadening the design experience for the team.

Independent Learning Experiences : 

1. Homework Assignments (weekly),
2. MATLAB Exercises in image and speech processing (weekly),
3.Testing (Preliminary Exam, Mid-Term Exam, Final Exam)

Contribution to the Professional Component: 

(a) College-level Mathematics and Basic Sciences: 1.25 credit hours
(b) Engineering Topics (Science and/or Design): 1.75 credit hours
(c) General Education: 0.0 credit hours
Total credits: 3

Prepared by: 
Lawrence R. Rabiner
Date: 
April, 2011