Non-negative least squares deconvolution method for mirror-ground beamforming

Abstract

In order to improve the acoustic source identification performance of beamforming when ground reflection exists, an array point spread function was derived and a corresponding non-negative least squares deconvolution method was given for a mirror-ground beamforming method. Simulations of a known acoustic source indicate that the given method is correct, and it could not only clear the acoustic source identification results effectively by improving the spatial resolution and attenuating the sidelobe interference, but it could also be superior to the conventional beamforming deconvolution method. On this basis, experiments were conducted to validate the correctness of the simulations and the effectiveness of the mirror-ground beamforming deconvolution method in practical application.