Chapter

Independent Component Analysis and Blind Signal Separation

Volume 3195 of the series Lecture Notes in Computer Science pp 726-733

Second-Order Blind Source Separation Based on Multi-dimensional Autocovariances

  • Fabian J. TheisAffiliated withInstitute of Biophysics, University of RegensburgDepartment of Electrical and Computer Engineering, Florida State University
  • , Anke Meyer-BäseAffiliated withDepartment of Electrical and Computer Engineering, Florida State University
  • , Elmar W. LangAffiliated withInstitute of Biophysics, University of Regensburg

Abstract

SOBI is a blind source separation algorithm based on time decorrelation. It uses multiple time autocovariance matrices, and performs joint diagonalization thus being more robust than previous time decorrelation algorithms such as AMUSE. We propose an extensioncalled mdSOBI by using multidimensional autocovariances, which can be calculated for data sets with multidimensional parameterizations such as images or fMRI scans. mdSOBI has the advantage of using the spatial data in all directions, whereas SOBI only uses a single direction. These findings are confirmed by simulations and an application to fMRI analysis, where mdSOBI outperforms SOBI considerably.