MIMO Instantaneous Blind Identification Based on Second-Order Temporal Structure

Blind identification is of paramount importance for well-known signal processing problems such as Blind Signal Separation and Direction Of Arrival (DOA) estimation. This paper presents a new method for Multiple-Input Multiple-Output Instantaneous Blind Identification based on second order temporal statistical variabilities in the data, such as non-whiteness and non-stationarity. The method basically consists of two stages. Firstly, based on certain assumptions about the statistical structure and diversity of the signals and system, and using subspace techniques, the problem is formulated in such a way that each column of the mixing matrix satisfies a system of multivariate homogeneous polynomial equations [1

• Type: Journal
• Year: 2008
• Where: TSP
• Authors: Jakob van de Laar, Marc Moonen, Piet C. W. Sommen
• Keywords: Blind Signal Separation, Instantaneous Blind, Multiple-Output Instantaneous Blind, TSP 2008