Real-time active noise control with preservation of desired sound

Abstract

In the active sound control problem a quite arbitrary bounded domain is shielded from noise generated outside by implementing secondary sound sources on the perimeter. The sound generated by interior sources, also known as desired sound, is supposed to be protected inside the shielded domain. If the desired sound is present, it is required to remain it unaffected by the control. This problem becomes much more complicated since the secondary sources have a reverse effect on the input data. In the current paper, a novel potential-based algorithm for active sound control is applied to attenuate noise with the preservation of desired sound. The uniqueness of this algorithm is that it realizes an active sound control via the utilization of surface potentials which have a projection property. The algorithm automatically removes the contribution of controls and desired sound from the total input field to be measured. The applicability of the algorithm is demonstrated via a series of numerical experiments. In addition, some factors such as the effect of the number of controls and sensors are also discussed.

Study of the nonlocal active sound control with preservation of desired field in time domain

“…The algorithm JASA is potentially applicable to realization in real-time regimes. It was implemented for the first time in the frequency domain by Zhou and Utyuzhnikov in (Zhou and Utyuzhnikov, 2020).…”

“…This happens because the level of noise attenuation depends on the frequency. It is clear that the higher frequency the lower level of noise attenuation occurs (Epain and Friot, 2007;Zhou and Utyuzhnikov, 2020).…”

“…The deterioration in the noise attenuation occurs because the operation of nonlocal control becomes rather local in case the external surface Γ − is too close to the boundary surface Γ. As demonstrated in(Zhou and Utyuzhnikov, 2020), a control point source operates locally in its some vicinity.…”

Optimized nonlocal active sound control in frequency domain

“…A practical algorithm for real-time implementation of NASC is proposed in [42]. Zhou and Utyuzhnikov have recently realized this algorithm in both frequency and time domains in [44,45], respectively, and carried out numerical experiments including sensitivity analysis. In particular, it is shown that in the case of sparse distributions of the sensors and controls, the nonlocal control can provide much greater noise attenuation than the local control even if the primary field is supposed to be available.…”

“…More precisely, this effect occurs if the desired sound component is present and essential enough. As noted in [44], with a sparse distribution of sensors and controls the projection property of the potentials can be corrupted.…”