Use of psychoacoustics models as non-destructive means for active noise control systems

Sound quality has always been a major and interesting concern, and more so in the acoustics of the automotive industry. With the new trend of vehicle weight reduction towards further fuel efficiency, the significance of the challenge of keeping low the noise level and vibration inside cars seems to be at its highest; as such noise can affect the car performance, as well as the comfort and general health of passengers. The noise sources and appropriate filtering, control and compensation techniques have been the subject of authors' other research works [1, 2], but equally important is the accurate assessment of passenger's perceived control efficiency through non-destructive means. In doing so, psychoacoustic metrics should be taken into account; and therefore an evaluation model for psychoacoustics sound quality is required. In this research paper, an attempt has been made to design and implement a multi-channel active noise control system (ANC), which can be employed for controlling multi-harmonic noises that are emitted from the engine inside a vehicle's cabin. In order to obtain accurately assessed results, the sound quality (SQ) is primarily evaluated through correlation between objective SQ metrics (Loudness, Sharpness, ...etc.) and subjective rating. Moreover, the newly developed correlation metrics (i.e. subjective pleasantness factor (SPF) and subjective annoyance factor (SAF)) is estimated to predict the response of an interior SQ influenced by noise in vehicle's cabin. These metrics is used to change the parameters of the developed ANC algorithm and so non-destructively controlling the corresponding SQ.