4 units – Winter Quarter
Lecture: 4 hours
Prerequisite: EEC 150B; STA 120 or MAT 131 or MAT 167 recommended.
Grading: Letter; homework (10%), midterm (20%), course project (30%), final exam (40%).
Theory and design of digital filters. Classification of digital filters, linear phase systems, all-pass functions, FIR and IIR filter design methods and optimality measures, numerically robust structures for digital filters.
Expanded Course Description:
This class is a core graduate level course in Digital Signal Processing (DSP) and is essential for students planning to pursue research in this area. The goal of this class is to provide an in- depth treatment of the topic of digital filter design. In specific, the first part of the course covers the theoretical aspects of the digital filter design problem whereas the second part addresses the implementation of these filters via numerically robust structures. A filter design project where students can experiment with the inherent filter design tradeoffs and pursue novel applications in data compression, communications and genomics to name a few, is a key component of this class. By the end of the term, we hope to provide a thorough and unified treatment of digital filters and their role in contemporary applications to the level where the student can engage in research in these areas.
STATEMENT OF COURSE DESIGN PROJECT:
The students are first introduced to the theoretical aspects of the digital filter design problem. They then study well established techniques for designing such filters and are presented with a generic framework for analyzing the performance of these filters when implemented with finite numerical precision. In the design project, students are encouraged to explore new applications and are therefore given the opportunity to use their acquired knowledge to optimize digital filters to best meet certain specifications and/or to minimize a specific error criterion. In these design problems, the student typically must choose between a number of alternatives in order to achieve the best performance under a set of constraints (implementation cost, error criterion, magnitude distortion, phase distortion, quantization effects ?. etc.). These design tasks involve both theoretical derivations and computer simulations and in almost all cases, do not have a unique solution.
ABET Category Content:
Engineering Science: 2 credits
Engineering Design: 2 credits
EEC150B and EEC151 cover some of the same topics, but at much different levels. As a graduate level course, the overlapping topics are addressed in much more detail.
Last revised: February 2006