Separación de señales usando análisis de componentes principales y muestreo compresivo con mediciones mínimas

To increase the efficiency of the independent component analysis (ICA) and reduce the computational complexity of the system, this paper proposes a methodology based on the compressive sampling with minimum measurements. This allows compressing and modifying the Gaussian characteristics of audio signals. ICA is one of the most widely used schemes for separating the sources involved in an audio mixture received in a set of sensors. However, for proper operation it is required that the signals involved in the mixture do not have Gaussian characteristics or at most only one of them be Gaussian, characteristics that are not satisfied by the audio signals, which have in general Gaussian characteristics. The proposed methodology allows obtaining a better separation with lower computational complexity, and with a more efficient separation of the mixed signals, even if the signals involved in the mixture are of Gaussian toe.