Artificial neural network algorithms for active noise control applications

This paper shows the use of several methods commonly applied to training Artificial Neural Networks (ANN) in Active Noise Control (ANC) systems. Although ANN are usually focused on off-line training, real-time systems can take advantage of modern microprocessors in order to use these techniques. A theoretical study of which of these methods suit best in ANC systems is presented. The results of several simulations will show their effectiveness.