The performance of common linear algorithms in active noise control applications can be degenerated mainly due to the unmodeled nonlinearities of loudspeakers as actuators in noise attenuation process. The aim of this article is to propose different methods such as prediction error method (PEM), nonlinear autoregressive network with exogenous input (NARX), and series-parallel NARX network based on neural network to experimentally identify the nonlinear behaviour of a loudspeaker. A model of loudspeaker is being used in noise cancellation and control algorithms; hence, its validity and robustness to input amplitude and frequency plays a crucial role in noise control. The results of this article, which are completely based on real and experimental data, demonstrate that neural network based series-parallel NARX model is the best estimator for the fully nonlinear behaviour of the loudspeaker. The capability and robustness of this estimator in comparison with other methods is examined by different test inputs with different amplitudes and frequencies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.