We introduce a new blind source separation approach, based on modified Kullback– Leibler divergence between copula densities, for both independent and dependent source component signals. In the classical case of independent source components, the proposed method generalizes the mutual information (between probability densities) procedure. Moreover, it has the great advantage to be naturally ext...

Blind source separation technology refers to the process for observing the recovery of source signals by mixed signals through statistical analysis on the characteristics of source signals under the situation that the source signals and signal transmission signals are unknown, which is applied in many fields, particularly used extensively in processing speech signals, array signals, images and ...

1 Summary The goal of blind deconvolution and source separation is to unravel the effects of an unknown linear transformation on a unknown signal source. For blind deconvolution, the transformation is a linear finite-impulse response (FIR) filter, and for blind source separation it is a matrix of mixing coefficients. A general architecture for these blind adaptive algorithms consists of an adju...

Nonnegative tensor factorization (NTF) of multichannel spectrograms under PARAFAC structure has recently been proposed by Fitzgerald et al as a mean of performing blind source separation (BSS) of multichannel audio data. In this paper we investigate the statistical source models implied by this approach. We show that it implicitly assumes a nonpoint-source model contrasting with usual BSS assum...

This paper concerns the blind separation of P complex convolutive mixtures of N statistically independent complex sources, with underdetermined or noisy mixtures i.e. P < N . Our approach exploits the assumed distinct statistical properties of the sources: P sources are non-stationary, while the others are stationary. Our method achieves the ”partial separation” of the P non-stationary sources....

In graph signal processing (GSP), prior information on the dependencies in is collected a which then used when or analyzing signal. Blind source separation (BSS) techniques have been developed and analyzed different domains, but for signals research BSS still its infancy. this paper, gap filled with two contributions. First, nonparametric method, relevant to GSP framework, refined, Cram\'{e}r-R...