Secondary source placement optimization for sound field reproduction assisted by sound field interpolation

Sound field reproduction is a method that uses a set of secondary sources to reproduce a predefined desired sound field within a target area. The placement of these secondary sources is a crucial factor affecting reproduction performance, while the number of obtainable desired sound pressure values within the target area is another limiting factor. To address the issues of an insufficient number of desired sound pressure data in the secondary source placement optimization problem, this paper proposes a sound field interpolation-assisted secondary source placement optimization method. This involves first utilizing sound field interpolation to obtain more desired sound pressure data within the reproduction area, and then optimizing the secondary source placement based on both measured and interpolated sound pressure values. This study uses both matching pursuit and simulated annealing as representative secondary source placement optimization methods, assisted by the equivalent sources and kernel interpolation as examples of sound field interpolation methods, to perform sound field reproduction in a two-dimensional simulation model and a public three-dimensional room measured dataset. The results demonstrate that our proposed method can significantly improve the performance of secondary source placement optimization methods, thereby enhancing sound field reproduction performance. This paper also investigates other influencing factors such as the number of secondary sources, the regularization parameters, and the room reverberation.