Statistical Signal Processing
Academic Study Board of the Faculty of Engineering
Teaching language: English
EKA: T450016102
Censorship: Second examiner: External
Grading: 7-point grading scale
Offered in: Odense
Offered in: Autumn
Level: Master
Course ID: T450016101
ECTS value: 5
Date of Approval: 31-08-2018
Duration: 1 semester
Version: Archive
Course ID
Course Title
ECTS value
5
Internal Course Code
Responsible study board
Administrative Unit
Date of Approval
Course Responsible
Programme Secretary
Offered in
Level
Offered in
Duration
Mandatory prerequisites
Learning objectives - Knowledge
The student will acquire knowledge about:
- Continuous-time and discrete-time state space models
- Auto-correlation function and cross-correlation function
- Power spectral densities
- Filtering of random signals
- Discrete-time Wiener filter and deconvolution
- Basic, extended and unscented Kalman filter
- Particle filters
- Non-parametric and parametric PSD estimation
- Ideal stochastic, non-parametric system identification
- System identification with extraneous noise on input and/or output measurements
- Steepest descent algorithms
- LMS adaptive filter and variants and their applications
- RLS adaptive filter and their applications
Learning objectives - Skills
The student is able to:
- Combine and apply methods from statistics and signal processing for advanced signal processing
- Analyse and estimate stochastic signals in time and frequency domain
- Analyse, design and use optimal recursive and adaptive algorithms for signal processing
Learning objectives - Competences
The student is able to handle:
- Modelling, analysis and processing of stochastic signals and noise
- Comparing and evaluating the applicability of statistical signal processing methods in specific applications
Content
- State space description of LTI systems
- Analysis of random signals and noise
- Optimal filtering
- Recursive and linearized signal (state) estimation
- Probabilistic signal (state) estimation
- Spectral estimation and system identification
- Adaptive filtering
URL for Skemaplan
Teaching Method
Number of lessons
48 hours per semester
Teaching language
Examination regulations
Exam
Name
Exam
Examination is held
In the end of the semester.
Tests
Exam
EKA
T450016102
Name
Exam
Description
A drawn question is the starting point of the examination; however, the examination can be in the total syllabus of the course, if it is relevant for the discussion
Form of examination
Oral examination
Censorship
Second examiner: External
Grading
7-point grading scale
Identification
Student Identification Card
Language
English
ECTS value
5