Search

A new learning algorithm is proposed to solve the separation problem of the blind nonlinear underdetermined mixtures. The mixing system is characterised by the post-nonlinear structure and concurrently the number of sensors is less than the number of sources. The proposed algorithm utilises the generalised Gaussian distribution to model the prior probability distribution of the source signals and the mixing variables. A novel iterative technique based on alternate optimisation within the Bayesian framework has been developed for estimating the source signals, mixing matrix and the nonlinear distortion. It is shown that through formal Bayesian derivation, the update of the mixing matrix can be decomposed into two separate constituents given by linear and nonlinear parts. Furthermore, the post-nonlinear distortion functions in the mixing model are approximated by a set of polynomials and the coefficients are found by solving a least square error problem. Simulations have been carried out to verify the effectiveness in separating signals under nonlinear and underdetermined conditions. An average margin of 130% improvement has been obtained when compared with the existing linear algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes a new nonlinear blind source separation algorithm with hybridisation of fuzzy logic based learning rate control and simulated annealing to improve the global solution search. Benefits of fuzzy systems and simulated annealing are incorporated into a multilayer perceptron network. Fuzzy logic control allows adjustments of learning rate to enhance the rate of convergence of the algorithm. Simulated annealing is implemented to avoid the algorithm becoming trapped in local minima. A simple and computationally efficient method for controlling learning rate and ensuring a global solution is proposed. The performance of the proposed algorithm in terms of convergence of entropy, is studied alongside other techniques of learning rate adaptation. Simulations show that the proposed nonlinear algorithm outperforms other existing nonlinear algorithms based on fixed learning rates. [ABSTRACT FROM AUTHOR]