A Multiple-Channel ANC With Neural Secondary-Path Model for Railway Train Systems

This paper propose a novel active noise control (ANC) approach by employing IIR filter and neural network techniques, which is suitable to effective interior noise reduction for the systems. We construct a multiple-channel IIR filter module which is a linearly augmented framework with a generic IIR model to create primary control signal. A three-layer perceptron neural network is employed for establishing a secondary-path model to represent air channels among noise fields. Since the IIR module and neural network is connected in series, the output of a IIR filter is transferred forward to the neural model to generate a final ANC signal. A gradient descent optimization based learning algorithm is analytically derived for optimal selection of ANC parameter vectors. Moreover, re-estimation of partial parameter vectors in the ANC system is proposed to realize online learning mechanism. Stability analysis of the proposed ANC system is achieved in which we demonstrate sufficient conditions of stability from a IIR filter module. Lastly, we carry out numerical study to test our ANC methodology with realistic interior noise measurement obtained from the Korean trains.