Prediction filter design for active noise cancellation headphones

Abstract

Digital active noise control (ANC) for headphones usually has to predict the noise because of the latency of common audio converters. In adaptive feedback ANC, the prediction is based on the noise that entered the headphone. This noise is lowpass filtered because of the physical barrier of the ear cups. In this study, this low-pass characteristic is exploited to define a prediction filter which does not require real-time updates. For broadband noises, the prediction filter performs better than adaptive prediction methods like the least mean squares algorithm or iterated one-step predictions in the relevant frequency band. This is shown in simulations as well as in measurements. In addition, the authors show that their prediction filter is more robust against changes in the acoustics of the headphone. © The Institution of Engineering and Technology 2013.