Feedforward active noise control using wavelet frames: simulation and experimental results

Wavelet frames are an appropriate alternative approach for signal approximation especially for noise/vibration signals. This paper presents an active noise control scheme using various wavelet frames with an innovative structure. In this structure, the authors employ a nonlinear static mapping cascaded with an infinite impulse response filter to model the dynamic part of the network. Online dynamic back propagation learning algorithms are applied based on the gradient descent method to adjust the network parameters. Simulations are carried out to compare the performance of filtered-x least mean square and filtered-x back propagation neural network algorithms with the proposed method. Experiments are then designed and conducted to evaluate the developed algorithms. Both simulation and experimental results show the superior performance of the proposed method in terms of a fast convergence rate and noise attenuation while avoiding the curse of dimensionality.