Robust estimation of open aperture active control systems using virtual sensing

Abstract

For active noise control systems in an open aperture application, virtual sensing system is often needed to overcome design constraint in terms of microphone placement. The virtual sensing system, however, make assumptions to the acoustical field and thus is highly sensitive to any changes in the primary field. Taking the window aperture ANC application for example, incoming disturbance signals could impinge from multiple locations, altering the spatial correlation between the physical and virtual microphones. For instance, in the context of a high-rise apartment window, aircraft noise would propagate downwards from the sky, whereas traffic noise propagate upward from the ground. This paper considers the estimation performance of the remote microphone technique, an example of a virtual sensing method, in an open aperture application when faced with changes in the primary noise source. Additionally, this paper introduces a new method for estimating the spectral density due to multiple sources through the incoherent decomposition method. It has been shown experimentally that this algorithm is able to improve the nearfield estimation of the system.