Course Catalog 2012-2013
Basic

Basic Pori International Postgraduate Open University

|Degrees|     |Study blocks|     |Courses|    

Course Catalog 2012-2013

SGN-2607 Statistical Signal Processing, 6 cr

Additional information

Course website: http://www.cs.tut.fi/courses/SGN-2607/
Suitable for postgraduate studies

Person responsible

Heikki Huttunen

Lessons

Study type P1 P2 P3 P4 Summer Implementations Lecture times and places
Lectures
Excercises
Assignment



 



 
 2 h/week
 2 h/week

+2 h/week
+2 h/week
 20 h/per



 
SGN-2607 2012-01 Tuesday 12 - 14, TB222
Tuesday 12 - 14, TB215

Requirements

Final examination, weekly exercises and a Matlab assignment.
Completion parts must belong to the same implementation

Learning outcomes

After passing this course the student will understand what statistical parameter estimation means, and how the methods can be applied in signal processing. Additionally, the basics of statistical detection theory will be covered.

Content

Content Core content Complementary knowledge Specialist knowledge
1. Basic concepts of estimation. What is an estimator and how to compare their performance. Cramer-Rao lower bound for estimator variance.     
2. Classical tools for estimation of a deterministic parameter: minimum variance unbiased estimator (MVUE), sufficient statistic, best linear unbiased estimator (BLUE), least squares estimator.     
3. Sparse and regularized estimators: Ridge regression and the LASSO. Design of sparse classifiers via regularized logistic regression.     
4. Estimation of nondeterministic parameters: the Bayesian approach.     
5. Basic principles of detection theory: Likelihood ratio test, ROC curve.     

Evaluation criteria for the course

The final grade comes from the final exam. The grade is incremented by one if at least 50% of weekly exercises are done.

Assessment scale:

Numerical evaluation scale (1-5) will be used on the course

Partial passing:

Completion parts must belong to the same implementation

Study material

Type Name Author ISBN URL Edition, availability, ... Examination material Language
Book   Fundamentals of Statistical Signal Processing - Estimation Theory   Kay, S.M.   0-13-042268-1          English  
Lecture slides   Statistical Signal Processing   Heikki Huttunen            English  

Prerequisites

Course Mandatory/Advisable Description
MAT-33311 Tilastomatematiikka 1 Advisable   1
MAT-33317 Statistics 1 Advisable   1

1 . The contents of the two courses are equivalent.

Prerequisite relations (Requires logging in to POP)



Correspondence of content

Course Corresponds course  Description 
SGN-2607 Statistical Signal Processing, 6 cr SGN-2606 Statistical Signal Processing, 5 cr  
SGN-2607 Statistical Signal Processing, 6 cr SGN-21006 Advanced Signal Processing, 5 cr  

More precise information per implementation

Implementation Description Methods of instruction Implementation
SGN-2607 2012-01   Lectures
Excercises
Practical works
   
Contact teaching: 0 %
Distance learning: 0 %
Self-directed learning: 0 %  

Last modified12.03.2012