6533b830fe1ef96bd1297963

RESEARCH PRODUCT

Alternative Diagonality Criteria for SOBI

subject

description

Blind source separation (BSS) is a multivariate data analysis method, whose roots are in the signal processing community. BSS is applied in diverse fields, including, for example, brain imaging and economic time series analysis. In the BSS model there are interesting latent uncorrelated variables, and the aim is to estimate the latent variables from multiple linear combinations of them. In this article we assume that these variables are weakly stationary time series, and we consider estimation methods which are based on approximate joint diagonalization of autocovariance matrices. In the popular SOBI estimator, a set of matrices is most diagonal when the sum of squares of their diagonal elements is maximal. Here we investigate other criteria to measure the diagonality of matrices. Applying both asymptotic results and simulations, we will study how the use of different diagonality measures affects the separation performance. Also, a method to choose the measure optimally based on data is proposed.