Instructor: Martin Hagan
Office: 311 E.S. Phone: (405) 744-7340
Email: mhagan@okstate.edu
Lessons
in Estimation Theory for Signal Processing, Communications, &
Control
Author: - Mendel
Publisher: Prentice Hall, 1995
There are two main topics covered in this course: system identification and state estimation for discrete-time systems. System identification is a two-stage process for developing models of dynamic systems. The first stage is the determination of the system order, and the second stage is the estimation of the system parameters. This course will present several techniques for order determination, using autocorelation functions, partial autocorrelation functions and generalized partial autocorelation functions. We will also discuss a number of methods for parameter estimation, including least squares and maximum likelihood techniques. The course will also address applications of the developed models to problems in prediction, control, signal processing, and coding.
State estimation is the process of estimating dynamic hidden variables based on a set of related measurable variables. The principal technique for state estimation in linear systems is the Kalman filter (a generalization of the Weiner filter). In the second half of this course we will develop the Kalman filter in a simple step-by-step process. We will also discuss applications of Kalman filtering to problems in prediction, control, inertial navigation, etc.
Prerequisite: ECEN 5513 or equivalent background in random
systems
The course grade will be based on two examinations, homeworks and quizzes, and a term project. For the project you will apply the system identification techniques covered in the first half of the semester to a set of real-life data.
ECEN/MAE 5513, or equivalent background in probability.