Abstract
We propose a method for source separation of convolutive mixture based on nonlinear prediction-error filters. This approach converts the original problem into an instantaneous mixture problem, which can be solved by any of the several existing methods in the literature. We employ fuzzy filters to implement the prediction-error filter, and the ecacy of the proposed method is illustrated by some examples.
Highlights
The problem of blind source separation has attracted a lot of attention in the last years due to its applicability to many fields, such as digital communications, pattern recognition, and biomedical engineering [1].Since the pioneering work by Herault et al [2], a myriad of different techniques for source separation has been developed
A model of this kind bears a strong resemblance with the idea of convolution and, in the case of blind source separation, this is exactly the reason why it is usually designated by the name of convolutive mixture
The output signals of the nonlinear prediction error filters are used as inputs for the fastICA algorithm [1], in order to complete the separation process
Summary
The problem of blind source separation has attracted a lot of attention in the last years due to its applicability to many fields, such as digital communications, pattern recognition, and biomedical engineering [1].Since the pioneering work by Herault et al [2], a myriad of different techniques for source separation has been developed. In the great majority of the works, the mixing process is modeled as an instantaneous linear system. Even highly complex systems, such as those found in biomedical signal processing, can be surmised to suit a framework of this sort (see, for instance, [1, 3, 4] and references therein for more details on applications and reviews about the existing separation criteria). Some mixing systems are of a more complex nature, and, as a consequence, cannot be modeled as linear combination of the instantaneous source signals (which originates the notation instantaneous mixture). There are practical instances in which the measured signals must necessarily be understood as being formed by a combination of different sources and delayed versions of them. A model of this kind bears a strong resemblance with the idea of convolution and, in the case of blind source separation, this is exactly the reason why it is usually designated by the name of convolutive mixture
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
