Finnish, Course can be passed in English.
Spring, periods 4.
1. is able to utilize the generic linear model as a representation for parameter estimation2. can apply typical deterministic and random parameter estimation methods for different estimation problems3. is able to determine statistical properties of estimators and make comparisons between them4. can form a basic state-variable model and utilize Kalman filtering for state estimation5. is able to apply basic methods of detection theory for solving simple detection problems6. can implement the learned methods and assess their statistical properties with the Matlab software
This course provides basic knowledge of statistical signal processing, in particular, estimation theory and its applications in signal processing. Topics: 1. Introduction, 2. Modeling of estimation problems, 3. Least Squares estimation, 4. BLUE-estimation, 5. Signal detection, 6. ML estimation, 7. MS estimation, 8. MAP estimation, 9. Kalman Filter.
Face-to-face teaching and homework assignments.
Lectures (24 h), exercises (24 h) and Matlab homework assignments (20 h).
Computer Science and Engineering students and other Students of the University of Oulu.
031078P Matrix Algebra, 031021P Probability and Mathematical Statistics
521337A Digital Filters, 031050A Signal Analysis. These courses provide complementary information on digital signal processing and stochastic signals. The courses are recommended to be studied either in advance or simultaneously.
J. Mendel: Lectures in estimation theory for signal processing, communications and control, Prentice-Hall, 1995. M.D. Srinath, P.K. Rajasekaran, R. Viswanathan: Introduction to Statistical Signal Processing with Applications, Prentice-Hall, 1996, Chapter 3. Lecture notes and exercise material.
The course is passed with intermediate exams or final exam and accepted Matlab exercise.
Read more about assessment criteria at the University of Oulu webpage.
The course unit utilizes a numerical grading scale 1-5. In the numerical scale zero stands for a fail.