**3480
University Street, Montreal, Quebec, CANADA**

**McGill University**

Department of Electrical and Computer Engineering

Academic Integrity

McGill University values academic integrity. Therefore, all students must understand the meaning and consequences of cheating, plagiarism and other academic offences under the code of students conduct and disciplinary procedures (seeacademic integrityfor more information).

ECSE-621B STATISTICAL DETECTION AND ESTIMATION

Winter 2013

General Information:

Instructor:Prof. H. Leib, Tel. 398-8938, room MC757

email : harry.leib@mcgill.ca

office hours : Thursday 15:00 -16:30

Teaching AssistantTBD

TBD

TBD

Text book:H. Vincent Poor, An Introduction to Signal Detection and Estimation, 2'nd edition, Springer-Verlag 1994

References:Harry L. Van Trees, Detection, Estimation, and Modulation Theory, Part 1, Wiley 1968 (reprinted version)

Papers from technical journals

Final mark composition:Assignments (20%), one midterm (2h, open books, 30%), term project (50%)

Schedule/Location:Tuesday and Thursday, 11:35-12:55, room ENGTR2120

First Class:Tuesday, Jan. 8, 2013

Term project:The term projects are done by each student individually. The subject can be selected by the students ; however it requires the agreement of the instructor. The deliverable is a project report of 20-25 pages excluding figures and tables. Project Guidelines :

Milestones for term project:Submission of title and 300 words abstract - Feb. 5, 2013

Submission of final paper - April 9, 2013 after the lecture.

__Course Outline__

The subject of Statistical Detection and Estimation lies at the intersection of telecommunication systems engineering, signal processing,

and mathematical statistics. It provides analytical tools for the analysis and synthesis of telecommunication and signal processing systems.

This subject has many applications in other areas too, such as : control, computer science, and bioengineering. The main objective of this

course is to provide a solid foundation, enabling students to apply such statistical tools in their own research. This course covers the

following main topics:1) Classical detection and estimation theory:

- Hypothesis testing (Neyman-Pearson and Bayes criteria), Minimax hypothesis testing
- Estimation theory (maximum-likelihood, maximum a-posterior, minimum mean-square error), Performance bounds
- Composite hypothesis testing, Generalized likelihood
2) Applications for discrete time signals in noise:

- Detection of signals in noise
- Estimation of signal parameters
- Detection of signals with nuisance parameters
3) Estimation of discrete time signals:

- Linear filtering
- Wiener filtering
- Kalman filtering
4) Discrete representation of continuous-time signals:

- Signal space concepts
- Karhunen-Loeve expansion of random processes
5) Detection and estimation of continuous time signals in noise:

- Detection of deterministic signals in noise
- Detection of random signals in noise
- Estimator-correlator structures

Announcements

TERM TEST

Type of exam :

Open books, open notes, all calculators allowed

Portable computers are not allowed Documents containing solutions to problems in the text book are not allowed.Date and Time :

Monday, March 25, 2013, 14:35-16:25

Location :

EDUC 433 (Education building room 433)

Material :

1) Class lectures from the beginning of the term and until March 19, 2013

2) Text book (Poor): pp. 1-97, p.141, pp. 157-158, pp. 169-181

Useful Links

Virtual lab on probability and statistics