Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
563215 | Signal Processing | 2009 | 8 Pages |
Abstract
In this paper a blind multiple-input/multiple-output (MIMO) space–time equalizer is described, dedicated to convolutive mixtures when observations have been pre-whitened. Filters preserving space–time whiteness are paraunitary; a parameterization of such filters with plane rotations is proposed. Theoretical developments then lead to a numerical algorithm that sweeps all pairs of delayed outputs. This algorithm involves the solution of a polynomial system in two unknowns, whose coefficients depend on the output cumulants. Simulations and performance of the numerical algorithm are reported.
Related Topics
Physical Sciences and Engineering
Computer Science
Signal Processing
Authors
Sylvie Icart, Pierre Comon, Ludwig Rota,