Docstoc

EE60S Kalman and Levinson Filtering

Document Sample
EE60S Kalman and Levinson Filtering Powered By Docstoc
					Module Name                  Kalman & Levinson Filtering
Module Code                  EE618

Module Co-ordinator          Selim Solmaz
Department                   Electronic Engineering

Module Level                 5
Credit rating                7.5 ECTS credits

Pre-requisites               Basic knowledge of linear algebra, linear system analysis and probability
                             theory
Co-requisites                Linear algebra


Aims                          To develop an expertise in stochastic systems and filtering theory for
                               applications in estimation and control problems.

Learning Outcomes            At the end of this module the student will:
                              Get the basic mathematical background on stochastic processes and
                                 discrete time system analysis
                              Understand the projection theorem and use it to derive various concepts
                                 in control theory
                              Learn about the derivation and the applications of Kalman filters as
                                 optimal state estimators
                              Familiarize with Levinson filters and reduced order state estimation


Time Allowance for Constituent Elements

Lectures                                            36 hours
Tutorials                                            0 hours
Class Tests                                          0 hours
Homework Assignments (8 x 6hr)                      48 hours
Independent study                                   36 hours


Indicative Syllabus

      Projection Theorem

      Discrete time stochastic processes and state space models

      Least Squares

      Kalman filtering

      Wide sense stationary processes

      Spectral analysis and the spectral density

      The Levinson filter, inverse scattering and prediction

      Sinusoid estimation
Assessment Criteria

Homework Assignments (8)                           100 %

Penalties: Late submission of assignment will not, in general, be accepted.

Pass Standard and any Special Requirements for Passing Modules: In order to pass this module,
students must achieve at least 40% in combined continuous assessment components.

Requirements for Autumn Supplemental Examination: The continuous assessment mark is carried
forward to the Autumn examinations as there is no facility available for repeating the continuous
assessment components of the course.

Continual Assessment Results: Assignments will be corrected within two weeks after submission,
provided that this does not extend past the end of the semester. Solutions and corrected scripts will be
available for viewing upon request.


Assessment Philosophy

The homework assignments include each specific topic learned during the course to achieve the
aforementioned learning outcomes. By working on the assignments individually or in groups, students get
the extensive working experience of the topics covered in class. The final grade is based on continuous
assessment and is calculated from the cumulative grades of the individual assignments.


Course Text                     Complete lecture notes will be handed out.

References                   Any book on the topics covered can be used as a reference textbook
                             Some suggestions are:
                              Introduction to Random Signals and Applied Kalman Filtering, Brown &
                                Hwang, 3rd Edition, Wiley, 1997
                              Kalman Filtering, Grewal & Andrews, 2nd Edition, Wiley Inter-science
                                2001.


Programmes currently utilizing module

Masters in Electronic Engineering

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:43
posted:3/16/2010
language:English
pages:2