Catalog data:
System identification, sampling, discrete systems, probability theory, stochastic
processes, spectral density, estimation theory, identification of systems with stochastic
in- and outputs, measurements in open- and closed-loop. |
Courseyear:
4
Semester: 4/0/0/0
Hours p/w: 4
Other hours: 0
Assessment: Written
Assessm.period(s): 1, 2
(see academic calendar) |
Prerequisites:
wb2203, wb2204, wi380 |
Follow
up: wb2301/5, wb2403/5 |
Detailed
description of topics:
-
Introduction system identification
-
System description with deterministic inputs, bode
diagram
-
Sampled signals in frequency domain, Theorema of
Shannon, Parseval, Fast Fourier Transform
-
Discrete-time linear systems, z-transform
-
Probability theory, random variables
-
Stochastic processes, distribution and density
functions, ergodicity
-
Stochastic processes in frequency domain, spectral
density, coherency
-
Identification of systems with stochastic inputs
-
Estimation theory
-
Estimation of spectral densities and
transferfunctions
|
Course
material:
A. van Lunteren, J. Dankelman. Signaalanalyse,
Dictaat WbMT (in Dutch)
|
References
from literature:
Ljung L. System Identification Theory for the user.
Prentice Hall Inc. 1987, 519P
Soderstrom T., Discrete-time Stochastic
Systems-Estimation and Control, Prentice Hall, 335P
Therrien C.W. Discrete Random Signals and Statistical
Signal Processing, Prentice Hall, 727P
|
Remarks
(specific information about assesment, entry requirements, etc.): |
Goals:
Introduction to non-parametric system identification techniques using the theory of
stochastic processes in time and frequency domain.
The courses wb2301 (Systemidentification A, non-parametric) and wb2403
(Systemidentification B, parametric) will further explain the practical use of this
theory. |
Computer
use: |
Laboratory
project(s): |
Design
content: |
Percentage
of design: 0% |