Abstract
A novel sound field reconstruction technique based on a time-domain Nearfield Equivalence Source Imaging (NESI) approach is proposed to identify locations and strengths of noise sources. Due to its time-domain nature, many problems of the conventional FFT-based methods such as spatial aliasing problem can be avoided. Tikhonov regularization that alleviates the ill-posedness inherent in the inverse problem with the least-squares optimization is employed to design multichannel time-domain inverse filters. A windowing design is also suggested to cope with the defocusing problem. The DOR is selected according to the condition number of the propagation matrix. In order to avoid singularity problems during processing, focus points are retreated away from the reconstruction surface. Post-processing is carried out using the time-domain filters to calculate acoustical variables including sound pressure, particle velocity, and active intensity. Furthermore, new features including the non-planar array, farfield reconstruction, and field interpolation/extrapolation are developed to enhance the present technique. Simulations and experiments are conducted to justify the proposed NESI technique using a two-dimensional (2D) array. Indicated in the results, the NESI technique proved effective in identifying sources of many kinds, including broadband, narrowband, stationary, and transient sources.