Bilinear blind source separation of EEG recorded simultaneously on several subjects

Doctoral student: 
Florent BOUCHARD
Supervisors: 
Name: 
Marco CONGEDO
Laboratory: 
GIPSA-lab
Name: 
Jérôme MALICK
Laboratory: 
LJK & Inria
Summary: 

Over the past 30 years Blind Source Separation (BSS) has established itself as a core methodology for the analysis of data in a very large spectrum of engineering applications.
This thesis aims at extending linear BSS models into bilinear models. When applied to electroencephalographic data, such extension can be used to exploit both spatial and temporal commonalities in the signal.