Fast Approximation of Nonlinearities for improving inversion algorithms of PNL mixtures and Wiener systems

摘要

This paper proposes a very fast method for blindly approximating a nonlinear mappingwhich transforms a sum of random variables. The estimation is surprisinglygood even when the basic assumption is not satisfied.We use the method for providinga good initialization for inverting post-nonlinear mixtures and Wiener systems.Experiments show that the algorithm speed is strongly improved and the asymptoticperformance is preserved with a very low extra computational cost.