Ricerc@Sapienza

Nonlinear acoustic echo cancellation (NAEC) can be mainly addressed by solving two different sub-problems: the estimation of the acoustic impulse response and the modeling of the nonlinearities rebounding in it, mostly caused by the electroacoustic chain. Both the modeling processes share an important characteristic: the majority of the parameters to be estimated are very close to zero, with only a small fraction of them having non-negligible magnitude. In this paper, a novel NAEC model is proposed taking into account both the above sub-problems under a joint optimization problem.

Linear-in-the-parameters nonlinear adaptive filters often show some sparse behavior due to the fact that not all the coefficients are equally useful for the modeling of any nonlinearity. Recently, proportionate algorithms have been proposed to leverage sparsity behaviors in nonlinear filtering. In this paper, we deal with this problem by introducing a proportionate adaptive algorithm based on an ℓ1-norm penalty of the cost function, which regularizes the solution, to be used for a class of nonlinear filters based on functional links.