Welcome to the home page for ECE531 "Principles of Detection and Estimation Theory" for Spring 2009.
announcements
- [02-Mar-2009] Please bring a calculator to the midterm exam on Thursday. I know the syllabus says that calculators are not permitted in the exams, but I am making an exception for the midterm on Thursday.
- [22-Jan-2009] An email was sent to the class mailing alias ece531@ece.wpi.edu today. If you did not receive this email, please contact Prof. Brown.
general
The required course textbook is An Introduction to Signal Detection and Estimation, 2nd edition, by H. Vincent Poor.
The course syllabus (pdf format) including expected course outcomes, grading information, and late policies.
ECE531 academic honesty policies.
ECE531 students with disabilities statement.
lecture notes and handouts
- Lecture 1: Introduction and review essential probabilistic concepts. (Corrected slides posted on 23-Jan-2009).
- Lecture 2a: A mathematical model for hypothesis testing and Lecture 2b: Bayesian hypothesis testing.
- We will spend roughly the first hour of Lecture 3 finishing up Bayesian hypothesis testing and then we will cover Lecture 3: Minimax hypothesis testing. (Corrected slides posted on 06-Feb-2009)
- Lecture 4a: Neyman-Pearson hypothesis testing and Lecture 4b: Detection of deterministic signals.
- Lecture 5: Composite hypothesis testing. (Corrected slides posted on 20-Feb-2009 and posted again on 03-Mar-2009). Also please see this explanation clarifying some of the details of the Bayes composite HT problem (especially slide 13).
- Lecture 6: Detection of discrete-time signals with random parameters. (Corrected slides posted on 03-Mar-2009).
- Lecture 7: Bayesian estimation and an introduction to non-random parameter estimation.
- Lecture 8: Non-random parameter estimation. Corrected slides uploaded on 19-Mar-2009.
- Lecture 9: Information inequality and the Cramer-Rao lower bound. Corrected slides uploaded on 27-Mar-2009.
- Lecture 10a: Maximum-likelihood estimation and Lecture 10b: Dynamic Parameter Estimation: System Model.
- Lecture 11: Dynamic Parameter Estimation: The Kalman-Bucy Filter.
- Lecture 12: Linear Estimation and Causal Wiener-Kolmogorov Filtering. Corrected slides (just slide 12 changed) uploaded on 19-Apr-2009.
- Lecture 13: Sequential Detection of Discrete-Time Signals. Also, course evaluations will be distributed in this lecture.
homework and solutions
There will be 10 homework assignments in ECE531, each worth 20 points. The lowest two scores will not be counted.
- Homework 1. Due by 8:50pm on 29-Jan-2009. Solution (corrected version posted on 3-Feb). Mean=16.9 Median=18, Max=20.
- Homework 2. Due by 8:50pm on 05-Feb-2009. Solution. Mean=15.8, Median=16.5, Max=20.
- Homework 3. Due by 8:50pm on 12-Feb-2009. Solution. Mean=16.3, Median=16.5, Max=20.
- Homework 4. Due by 8:50pm on 19-Feb-2009. Solution. Mean=17.2, Median=18.0, Max=20.
- Homework 5. Due by 8:50pm on 26-Feb-2009. Solution. Mean=15.6, Median=15, Max=20.
- There is no homework due on 05-Mar-2009.
- There is no homework due on 12-Mar-2009.
- Homework 6. Due by 8:50pm on 19-Mar-2009. Solution. Mean=17.6, Median=18, Max=20.
- Homework 7. Due by 8:50pm on 26-Mar-2009. Solution. Mean=18.1, Median=19, Max=20.
- Homework 8. Due by 8:50pm on 02-Apr-2009. Solution. Mean=18, Median=19, Max=20.
- Homework 9. Due by 8:50pm on 09-Apr-2009. Solution. Mean=17.4, Median=18, Max=20.
- Homework 10. Due by 8:50pm on 16-Apr-2009. Solution. Mean=17.9, Median=18, Max=20.
examinations and solutions
- A 90 minute midterm exam worth 340 points will be held from 6:00-7:30pm on 05-Mar-2009. Midterm exam and solution. Mean=234, Median=230, Max=340.
- A 150 minute comprehensive final exam worth 500 points will be held from 6:00-8:30pm on 23-Apr-2009. Final exam and solution. Mean=240, Median=235, Max=480.