EE 563: Introduction to Estimation Theory

Course Objectives and Description: In this course you will learn famous and frequently used parameter and state estimation techniques and algorithms that are widely used in many fields. The course will cover: least squares, best-linear unbiased estimation (BLUE), maximum-likelihood, mean-squared, maximum a posteriori, mean-squared state-prediction, -(Kalman) filtering, and -smoothing, extended Kalman filtering, and Unscented Kalman filtering. The course will also cover an overview of higher-order statistics. It not only will present derivations and performance analyses of the major estimation algorithms that are in use today, as well as their applications, but will also explain how and when many of the algorithms are interrelated.

Prerequisites: A graduate course in random processes in engineering (EE 562a).

Textbook: Lessons in Estimation Theory for Signal Processing, Communications and Control, J. M. Mendel, Prentice-Hall, New Jersey, 1995. An Errata is on-line.

Additional Readings:
o S. J. Julier and K. J. Uhlmann, "Unscented Filtering and Nonlinear Estimation," IEEE Proc., vol. 92, pp. 401-422, March 2004.
o E. A. Wan and R. van der Merwe, The Unscented Kalman Filter," in Kalman Filtering and Neural Networks, S. Haykin (Ed.), pp. 221-280, John Wiley, 2001.
o S. J. Julier and K. J. Uhlmann, "A General Method for Approximating Nonlinear Transformations of Probability Distributions," Tech. Report RRG, Dept. of Engineering Science, Univ. of Oxford, Nov. 1996.





Return to Contents?