On the blind source separation of nonlinear mixtures

In this paper we analyze the method proposed by Ehsandoust et. al for the blind source separation of nonlinear mixtures. Interestingly, the initially proposed method based on deriving the observed signals, performing an adaptive linear blind source separation, smooth the coefficients of the separation matrices and integrate the solution can be reduced, in some cases, by simply performing an adaptive linear blind source separation and smooth the coefficients of the separation matrices. Also, we extend the results for different sets of signals such as autoregressive signals and propose an alternative method, based on a General Regression Neural Network, for the smoothing of the Jacobian matrix.