EE 351M: Digital Signal Processing
Tuesdays & Thursdays 11-12:15PM ECJ 1.204 Unique Number 16300
Signal processing is rich with tools that have applications in a broad class of problems including communications, controls, image compression, sonar, radar, array processing, and digital video. The theory is both elegant and beautiful.This course provides a thorough treatment of DSP including the fundamental theorems and properties of discrete-time linear systems, filtering, sampling, and discrete-time Fourier Analysis.Questions that we will answer along the way:
- How do you sample continuous-time signals?
- How can you perform discrete-time processing of continuous-time signals?
- How can you perform continuous-time processing of discrete-time signals?
- What are the best discrete-time filters?
- What is the effect of finite-precision on numerical implementation?
- What is the relationship between the Fourier transform, Fourier Series, Discrete Fourier Transform, and the Fast Fourier Transform?
- How do we use the Discrete Fourier Transform to analyze potentially continuous time signals?
- The emphasis in the class will be on digital signal processing algorithms, their derivations, and of course their applications.
An innovative feature of this course is the final term project. This project is intended to be a hands-on creative learning experience. You will be expected to demonstrate your grasp of the material drawn from lectures, the text, or other sources, by applying digital signal processing techniques to some problem of interest to you. This will improve your ability to apply knowledge of mathematics, science, and engineering, as well as help you learn to identify, formulate, and solve engineering problems. There will be a final presentation competition and a written term paper. There will be cash awards for the two best project presentations and an additional prize for the overall winner.
New for Spring 2009, this course will discuss audio applications of digital signal processing. Of particular interest will be audio applications to create digital audio effects (drawing on examples from the DAFX book by Udo Zölzer). Ever wonder how to create that concert hall effect? What about that cool wah wah sound? Or sounds like echo, slapback, or flanging? Or pitch shifting (harder than you think)? We will also discuss speech processing (as an application of stationary signal processing) as well as practical applications like noise cancellation. Audio examples will be used in the course as well as in MATLAB programming assignments.
By the end of this course, you should be able to analyze discrete-time systems by examining their input and output signals. You should be able to compute a system output in either time or frequency given the system input and a description of the system. You should know the correct way to derive a discrete-time signal from a continuous-time signal and the conditions for perfect reconstruction. You should understand the difference between discrete-time signals and digital signals and the practical considerations. You should be able to filter continuous-time signals in discrete-time and vice versa. You should be able to design good digital filters based on analog filter designs that you already know. You should be able to devise optimal filters especially for stochastic signals and adaptive filters. You should be able to use the Fourier transform, Fourier series, Discrete Fourier transform, and the Z-transform and know when to use them. You should know a fast way to compute the Discrete Fourier transform. You should know how to analyze the spectrum of real signals – both deterministic and stochastic. You should be able to provide details examples of at least five different applications of discrete-time signal processing to electrical engineering.
Discrete-Time Signal Processing, (2nd Edition) by Alan V. Oppenheim, Ronald W. Schafer, and John R. Buck ISBN: 0-13-754920-2 Publisher: Prentice Hall Copyright: 1999 Published: 12/31/1998 [Required].
DAFX – Digital Audio Effects Edited by Udo Zölzer ISBN: 0-471-49078-4 John Wiley & Sons, 2002 [Optional].
Electronic Course Site
The course material is usually available in Blackboard. The site includes a weekly outline, lecture notes, project details, homework assignments, discussion groups, and chat rooms.
MATLAB in EE 351M DSP
MATLAB is a popular software package for signal processing. Matlab will be used as part of several homework assignments.
Signal Processing at UT Austin
- Wireless Networking and Communications Group
- Embedded Signal Processing Laboratory
- High-Order Statistical Signal Processing Laboratory
- Laboratory for Image and Video Engineering
- Laboratory for Vision Systems
Signal Processing Links
- SPLIB: Signal Processing url Library
- IEEE Signal Processing Society
- Signal Processing Conferences and Workshops