Elec 531: Digital Signal Processing

Fall 1998 - Rice University

Instructor	Aria Nosratinia, 2019 Duncan Hall, Ext. 5056, aria@ece.rice.edu
Time	M-W-F 10:00-11:00am
Place	Room A126 Abercrombie Lab
Text	Proakis and Manolakis, Digital Signal Processing, Prentice Hall McClellan et al., Computer Based Exercises for Signal Processing, Prentice Hall
References	Oppenheim and Schafer, Discrete-Time Signal Processing Rabiner and Gold, Theory and Application of Digital Signal Processing Mitra, Digital Signal Processing
Grading	Midterm exam (%20), Homeworks (%20), Final Exam (60%)
Prerequisite	One semester of undergraduate signals and systems. An undergraduate digital signal processing course is highly recommended, but not required.
Office Hours	Fridays 11:00am-12:00pm

This is a graduate level course in Digital Signal Processing (DSP). An undergraduate course in signals and systems is prerequisite, and the background of an undergraduate digital signal processing course is strongly recommended, although not required. This course focuses on the fundamentals of digital signal processing, and includes in-depth treatment of digital filter design and various FFT algorithms in the first 2/3 of the semester. The last 1/3 of the semester is dedicated to more advanced topics, where the students are introduced to multirate signal processing, filter bank theory, and adaptive filtering theory.

Overview
- Motivation
- Discrete-time signals
- Discrete-time systems
- Convolution
- The Z-transform
- Frequency response and the DTFT
- The Discrete Fourier Transform
Fast Fourier Transforms
- Computational motivation
- Decimation in time FFT
- Decimation in frequency FFT
- Matrix view of FFT
- Split-radix FFT
- FFT for composite N
- Chinese Remainder Theorem and Prime Factor Algorithm
- Goertzel and Winograd Algorithms
FIR Filter Design
- Causality issues in filter design, Hilbert Transform
- Generalized linear phase and filter symmetries
- Windowed filter design
- Frequency Sampling filter design
- Generalized frequency sampling design - L2 FIR filters
- Chebychev approximation and the alternation theorem
- Minimax FIR design and Remez Exchange Algorithm
IIR Filter Design
- Impulse invariance
- Bilinear mapping
- Summary of analog filter design
- L2 Design - Pade' Approximation
- Prony and Shanks methods
Multirate Signal Processing
- Applications - motivation
- Decimation and interpolation
- Sample rate conversion for rational D/I
- Approximate sample rate conversion for irrational D/I
- Polyphase implementation of sampling rate conversion
- DFT filter banks
- General filter banks - alias cancellation and perfect reconstruction
Adaptive Signal Processing
- Applications - motivation
- Wiener filtering
- The Widrow LMS algorithm
- Performance analysis of LMS algorithm
- Introduction to the RLS algorithm

Aria Nosratinia

Last modified: November 30, 1998

Elec 531: Digital Signal Processing

Fall 1998 - Rice University

Contents: