Valid from: Autumn 2014
Decided by: FN1/Anders Gustafsson
Date of establishment: 2015-03-27
Division: Electrical and Information Technology
Course type: Course given jointly for second and third cycle
The course is also given at second-cycle level with course code: EITN55
Teaching language: English
The course gives basic knowledge in statistical signal processing and treats the theory of independent and principal components, together with applications in signal separation. In most areas where sensor systems are used, the received signals contain undesired components or several superposed useful signals, which affects the transmission of information negatively. The course in Signal Separation deals with methods to separate the received signals in a general context, only using the information contained in the received signals. The traditional approaches to analyse, filter, compress and separate a combination of signals by means of second order statistics (e.g. correlation based methods) are extended to include higher order statistics (e.g. higher than second order moments). This leads to the concept of independent components in contrast to principal components.
Knowledge and Understanding
For a passing grade the doctoral student must
Competences and Skills
For a passing grade the doctoral student must
Judgement and Approach
For a passing grade the doctoral student must be able to comprehend literature as well as standards in this area
The following items are treated in the course: linear representation of multivariate data, random vectors and independence, higher order moments, gradients and optimization, learning rules for non-constrained and constrained optimization, estimation theory for signal separation, methods of least-squares and maximum likelihood, information theory, entropy cumulants, definition of Independent Components (ICA) and Principal Components (PCA), differences and similarities between PCA and ICA, methods for estimation of ICA: ICA by maximization of non-Gaussianity, ICA by maximum likelihood estimation, ICA by minimization of mutual information, ICA by nonlinear decorrelation and nonlinear PCA. Applications: acoustic signal separation and deconvolution, feature extraction from multivariate data, artifact identification from Electroencephalography (EEG) and Magnetoencephalography (MEG), prediction of time series data by using ICA.
Hyvärinen, A., Karhunen, J. & Oja, E.: INDEPENDENT COMPONENT ANALYSIS. J. Wiley, 2001. ISBN 047140540X.
Types of instruction: Lectures, laboratory exercises, exercises
Examination formats: Written exam, written assignments.
Two compulsory hand-in tasks are part of the examination
Grading scale: Failed, pass
Examiner:
Admission requirements: - ESS040 Digital signal processing OR ETI265 Signal processing in multimedia OR EITF15 Digital signal processing - theory and applications, or an equivalent course
Assumed prior knowledge: Fundamental mathematics, statistics and stochastic processes
Course coordinator: Nedelko Grbic nedelko.grbic@eit.lth.se
Course coordinators:
Web page: http://www.eit.lth.se/index.php?ciuid=868&L=