Blind signal separation

Blind signal separation: "Blind Channel estimation

Blind Signal Separation (BSS) is a recurrent problem in multi-sensors applications where observed signals can be modelled as convolutive or memoryless linear mixtures of unknown source signals. The aim of BSS is to retrieve source signals from observations by exploiting the available a priori information on the pdf of the sources. Real-time applications call for BSS methods that are computationally affordable, fast convergent, stable and reasonably accurate."