Footnote:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Description:
Let a high-dimensional random vector $\vec{X}$ can be represented as a sum of two components - a signal $\vec{S}$, which belongs to some low-dimensional subspace $\mathcal{S}$, and a noise component $\vec{N}$. This paper presents a new approach for estimating the subspace $\mathcal{S}$ based on the ideas of the Non-Gaussian Component Analysis. Our approach avoids the technical difficulties that usually exist in similar methods - it doesn't require neither the estimation of the inverse covariance matrix of $\vec{X}$ nor the estimation of the covariance matrix of $\vec{N}$.