Blind Deconvolution / EqualizationUsing State - Space

In this paper we propose extension of multichannel blind equalization problem assuming that both mixing and demixing models are described by stable linear state-space systems. The problem is formulated as an optimization task. New learning algorithms are developed which can be considered as extension of existing algorithms. By applying demixing state-space model we will be able to reduce complexity of separating systems in some practical implementations.