Neuro-adaptive active control of noise

This paper presents the development of an intelligent active noise control framework using neural networks. An active control system is designed utilising a feedforward control structure for optimum cancellation of broadband noise. The controller design relations are formulated such that to allow online design and implementation and, thus, yield a self-tuning control strategy. Neural networks are used at the modelling and control contexts and thus incorporated into the control strategy. Two alternative neuro-adaptive active control algorithms are proposed on the basis of this approach. The algorithms thus developed are tested and verified in the cancellation of broadband noise in free-field. (3 pages)