Multi-domain active sound control and noise shielding

Abstract

This paper describes an active sound control methodology based on difference potentials. The main feature of this methodology is its ability to automatically preserve "wanted" sound within a domain while canceling "unwanted" noise from outside the domain. This method of preservation of the wanted sounds by active shielding control is demonstrated with various broadband and realistic sound sources such as human voice and music in multiple domains in a one-dimensional enclosure. Unlike many other conventional active control methods, the proposed approach does not require the explicit characterization of the wanted sound to be preserved. The controls are designed based on the measurements of the total field on the boundaries of the shielded domain only, which is allowed to be multiply connected. The method is tested in a variety of experimental cases. The typical attenuation of the unwanted noise is found to be about 20 dB over a large area of the shielded domain and the original wanted sound field is preserved with errors of around 1 dB and below through a broad frequency range up to 1 kHz.

Potential-based methodology for active sound control in three dimensional settings

“…In addition, the control outputs are supposed to be accurately separable from the input data. The experiments confirm that the potential-based ASC method, validated in one-dimensional conditions, 22,25 can be extended to cover full three-dimensional acoustic conditions and achieve global noise cancellation while preserving the wanted sound.…”

“…22 One of the main reasons can be found in the design of the experimental model. That is, the control sources themselves cause disturbances to the sound fields.…”

“…22 Fig . 6(d) shows the similarity between the original wanted sound pressure-and the result • when the controls are switched on.…”

“…22,27 In contrast to standard techniques, this approach allows the existence of wanted sound in the protected domain.…”

“…The one-dimensional effectiveness of the method has already been demonstrated in our previous publication. 22 The number of control sources is kept to four in the numerical simulations as that coincides with the number of controls used in the experiment in Sec. IV.…”