(aside image)

A collection of ICA and BSS links

Software and data sets

For our software, see the subpage "Software" above.

The ICA Central held by J-F. Cardoso contains data sets and algorithms provided by the researchers working in these fields. Furthermore, it contains a link to a mailing list by which you can reach many researchers working in these fields.

ICALAB contains Matlab implementations of many ICA algorithms and blind source separation algorithms based on second-order temporal statistics.


"Independent Component Analysis" by A. Hyvarinen, J. Karhunen, and E. Oja, Wiley 2001. During the years, this our 500-page book has become a standard reference on independent component analysis and its applications and extensions. It consists of four parts: Mathematical preliminaries, Basic independent component analysis, Extensions and related methods, as well as Applications of ICA.

NEW! Advanced book "Handbook of Blind Source Separation, Independent Component Analysis and Applications", edited by P. Comon and J. Jutten, published by Academic Press, 2010. This handbook is intended as a reference for the researchers working in these fields. It contains up-to-date review chapters on different types of ICA and BSS methods as well as on many their extensions and applications.

"Independent Component Analysis: A Tutorial Introduction" by J.Stone, The MIT Press, 2004. A tutorial introduction to independent component analysis intended mainly for newcomers to the field.

"Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications" by A. Cichocki and S.-I. Amari, Wiley 2002. A handbook type book containing many ICA and BSS algorithms and approaches.


The most important journals publishing articles on ICA and BSS methods are:

IEEE Transactions on Signal Processing
Signal Processing
IEEE Transactions on Neural Networks
Journal of Machine Learning Research
Neural Computation
Neural Processing Letters
Neural Networks


The first workshop (conference) on ICA and BSS was held in Aussois, France, in January 1999. After that, these conferences have been arranged about every 1.5 years under slightly varying names. In 2010, the name of this conference was changed to the 9th Int. Conf. on Latent Variable Analysis and Signal Separation (LVA/ICA 2010) to correspond better to recent developments. Basic ICA and BSS methods are today well understood, and research has largely moved to related topics such as non-negative matrix factorizations etc. Regrettably, there is no homepage where the papers of the ICA99 workshop would be available. The papers of the later ICA workshops are available via the following links:

LVA/ICA 2010

Other conferences publishing papers on ICA and BSS include the IEEE Machine Learning for Signal Processing (MLPS) workshops. The latest of them, MLSP2010 was organized by us in Lapland. Formerly the name of this workshop was Neural Networks for Signal Processing. Also the signal processing conferences ICASSP and EUSIPCO, various neural networks conferences and many smaller conferences contain papers on ICA and BSS.


The link page of our ICA book is partly out-of-date.

Figure caption

The Computer Science and Engineering house where our department is located.