Syllabus (Spring 04)

Course Title: ECE746 Statistical Signal Processing

Class Meetings: MWF 11:15-12:05 at ELAB 327

Instructor: Professor Joon Ho Cho

Office) Marcus 215-A
Phone) (413) 545-4375
E-mail) jhcho@ecs.umass.edu  
Office Hours: MWF 9:05-9:55. 

Homepages: http://www.ecs.umass.edu/ece/ece746

 It is your responsibility to check every new announcement posted in the homepage. 

 Prerequisites:

ECE 603 (Probability and random processes), ECE 608 (Signal Theory), and computer programming skills (MATLAB or C). Contact the instructor for questions about prerequisites.

 Text:    

No textbook. We will jump over the reserved books in the library.

 References:

H. L. Van Trees, Detection, Estimation, and Modulation Theory, Wiley, 1971 

M.D. Srinath, R.K. Rajasekaran, and R. Viswanathan, Introduction to Statistical Signal Processing with Applications, Prentice Hall

S. M. Kay, Fundamentals of Statistical Signal Processing:Estimation Theory, Prentice Hall, 1993

S. M. Kay, Modern spectral estimation : theory and application, Prentice Hall, 1988

H. V. Poor, An introduction to signal detection and estimation 2nd ed, New York : Springer-Verlag, 1994

 Course Objective:

The course objective is to let the student obtain 

  1. the capability of identifying the engineering problems that can be put into the frame of statistical signal processing, 

  2. the capability of solving the identified problems using the standard techniques learned through this course, and

  3. the fundamental ideas of statistical signal processing that may help them study further and make significant contributions to the theory and the practice of statistical signal processing.   

 Course Requirements:

1st midterm exam (Date TBA, take-home):   15 %
2nd midterm exam (Date TBA, take-home):   20 %
In-class final exam (Date TBA, take-home):  25 %
Homework 30 %
Lists of Points of Confusion 10 %
100 %

Only approximately 1/3 of homework problems will be randomly selected and graded.  

Before first class meeting of each week, you are asked to submit a list of Points of Confusion. Each lists will be scored 0 or 1.  If you submit a list every week then you may get 10% of the total. 

Final grade will be based on total score. 

 Course Outline

I. Introduction 

II. Statistical Inference

Non-Bayesian detection and estimation
Bayesian detection and estimation
Convergence of a random sequence
Non-Bayesian detection and estimation
Bayesian detection and estimation
KL expansion, Sampling theorem, etc.
Non-Bayesian detection and estimation
Bayesian detection and estimation

III. Non-statistical Inference

Least squares
Methods of moments
Spectral estimation

 Grading Policies for Homework 

  1. Each (sub)-problem will be graded 100% (perfect), 60 % (good),  20 % (not enough), and 0%. For example, if a (sub)-problem is a 15-point problem, then the possible scores are 15, 9, 3, and 0. No other partial credit scores will be given.

  2. Late homework will not be accepted.

  3. Collaboration is allowed on the homework. However, students must submit his/her own solutions and provide the names of collaborators. 

  4. Scores will be normalized to assure proper weighting for each assignment. 

  5. Re-grade requests must be filed in writing within one week after the graded homework has been returned to students.

 Grading Policies for Exam 

  1. Each (sub)-problem will be graded 100% (perfect), 60 % (good),  20 % (not enough), and 0%. For example, if a (sub)-problem is a 15-point problem, then the possible scores are 15, 9, 3, and 0. No other partial credit scores will be given.

  2. Students may take exams earlier than the original schedule if he/she requests at least two weeks earlier. 

  3. Excused class absence (defined in Pre-registration Guide) of an on-campus student on an exam may lead to a make-up exam for the absentee.

  4. Re-grade requests must be filed in writing within one week after the graded exam has been returned to students.  

 Policy on Academic Dishonesty

If the instructor suspects academic dishonesty, the instructor will notify the student(s) and follow the procedure to report to the Academic Dishonesty Office without any exception. Students have the responsibility to be knowledgeable about 'What to know about academic dishonesty: a guide for students' issued by the University Ombuds Office.

 

1/26/04