**Blind Source Separation**

**Course Description**

This course discusses the models for source recovering through blind approaches. The problem arises in several different fields of study. In speech communication, a microphone in a room can simultaneously receive a mixture of several speech signals that need to be separated to render intelligible results. In biological studies, the sensors receive different electrical excitations where usually only one is of interest. In communication systems, the sensors in a receiving site collect data from different users which interfere in each other and then we need to separate them and identify each signal of interest. We discuss the mathematical modelling and the proposed solutions for such both cases of instantaneous and convolutive mixtures.

**Prerequisites**

Some knowledge of probability and stochastic processes and signal processing is required.

**Course Goal**

To provide to the students the ability to model the problem of blind source separation as well as the solutions for the cases of instantaneous and convolutive mixtures. The students are expected to be able, at the end of the course, to apply the methods of blind source separation in different scenarios and areas.

**Program / Syllabus**

- Motivation
- Second order statistics
- Principal Components Analysis (PCA)
- Independent Component Analysis (ICA)
- Measures of independence
- Strategies and algorithms
- Applications

**Textbooks**

- P. Comon (Editor), C. Jutten (Editor),
*Handbook of Blind Source Separation: Independent Component Analysis and Applications*, Academic Press, 2010.
- A. Cichocki, S. Amari,
*Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications*, John Wiley and Sons, 2005.
- Aapo Hyvärinen and Erkki Oja and Juha Karhunen,
*Independent Component Analysis*, Wiley, 2001.
- Simon Haykin (Editor),
*Unsupervised Adaptive Filtering, Volume 1: Blind Source Separation*. Wiley, 2000.

**Material (in Portuguese)**