Adaptive feedback active noise control using wavelet frames: simulation and experimental results

Two strategies for adaptive feedback active noise control (ANC) systems are presented here based on wavelet frames. In the first strategy, the noise at the cancellation point is estimated and used as the reference signal in the feedforward ANC system. In the second strategy only the error signal is utilized in the wavelet-based ANC system. The advantages of the second method are: simple structure and less computational load (no need to estimate noise) as well as less sensitivity to the accuracy of estimated secondary path filter. Its disadvantage is high sensitivity to learning rate. To treat this, an adaptive learning rate is proposed to maintain the stability and convergence rate simultaneously. Simulations are carried out for typical linear/nonlinear cases to compare the proposed methods with FxLMS and neural network-based ANC algorithms. Experiments are then conducted to evaluate the developed algorithms. Results show the superior performance of proposed methods in terms of the convergence rate and noise attenuation especially in the presence of an inaccurate identified secondary path filter.