Noise suppression using active noise control and masking.
Date of Issue2008
School of Electrical and Electronic Engineering
An innovative solution to make the public/workplace environment more pleasant by combining techniques of active noise control (ANC) and psychoacoustics is discussed in this thesis. A version of the psychoacoustic model-I, suitably modified for operational simplicity, has been deployed by the Integrated ANC-Masking (IANCM) system. The limitations of existing ANC systems, which result in an annoying residual noise, are overcome by applying a soothing and pleasant masker at the exact levels of loudness. A technique for selection of the right masker from a pool of desirable maskers has also been suggested, in addition to giving the end users a choice of selecting their own favorite maskers. Various adaptive algorithms have been compared under several critical factors prior to deciding their suitability for the current task. The system employs a pleasant sound for ascertaining the offline estimate of the secondary path, thereby providing the end-user a completely pleasant experience. The effects of varying the masker levels on the quality of output have been analyzed and compared with a conventional ANC system. Continual tracking of the residual noise enables the system to find the right levels of masker amplitudes. This enables the system to cater for a wide range of sources of unpleasant noise. To further improve the performance of the system, preprocessing the masker for dynamic range reduction has also been employed. In addition to this, statistical methods for clustering have been employed to provide a smooth transition between maskers, thereby avoiding a monotonous experience.
DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering