525.431 - Adaptive Signal Processing Course Homepage
Instructor Information
James Costabile
Email: jcostabile@ieee.orgHome Phone: (410) 707-7338
Course Information
Course Description
This course examines adaptive algorithms (LMS, sequential regression, random search, etc.) and structures (filters, control systems, interference cancellers), and properties and uses of performance surfaces. Adaptive systems are implemented as part of the course work. Problem exercises and a term project require computer use.
Prerequisites
525.427 Digital Signal Processing. Some knowledge of probability helpful.
Course Goal
To provide students with the ability to apply adaptive filtering techniques to real-world problems (e.g. adaptive interferrence cancellation, adaptive equalization) in order to improve the performance over static, fixed filtering techniques. To provide a theorectical basis of adaptive signal processing necessary for the students to extend their area of study to additional applications, and other advanced concepts in statistical signal processing.
Course Objectives
Acquire a "toolbox" of adaptive filtering techniques and understand the condistions in which they are most effective.
Provide the ability to determine the optimal filter and theoretical performance of adaptive filtering techniques in known environments
Provide the ability to understand and apply these techniques to simulated signals in an unknown environment in order to gain improvements in performance
When This Course is Typically Offered
This course is typically offered in the fall at APL.
Syllabus
Topics Covered
- Stochastic Processes
- Wiener Filters
- Linear Prediction
- Steepest Descent
- LMS / NLMS Adaptive Filters
- Performance: Steady State, Tracking, Transient
- Frequency Domain / Subband Filters
- Least Squares
- Recursive Least Squares
- Kalman Filters
- Blind Deconvolution / Equalization
Student Assessment Criteria
| Homework | 10% |
| Mid-term Exam (take-home) | 30% |
| Final Exam (take-home) | 30% |
| Project | 30% |
Computer and Technical Requirements
Familiarity with MATLAB is desirable.
Participation Expectations
It is expected that students will share real-world problems they face at work in order for the class to discuss how we could apply adaptive filtering techniques to solve these problems.
Textbooks
Textbook information for this course is available online through the MBS Direct Virtual Bookstore.
Course Notes
There are no notes for this course.
(Last Modified: 08-08-2009 at 1:24:08 PM)