Abstract
Far-field acoustic imaging algorithms using an optimized random array are investigated in this paper. Global optimization techniques including the Simulated Annealing (SA) algorithm and the Intra-Block Monte Carlo (IBMC) algorithm enables efficient search for optimal position and weights in microphone deployment. In addition to the commonly used delay-and-sum (DAS) beamformer, the minimum variance distortionless response (MVDR) algorithm, the Equivalent Source Inverse Filtering (ESIF) algorithm and the Time Reversal (TR) algorithm are suggested to reconstruct and visualize the sound field. These far-field imaging algorithms are implemented on a 30-channel optimized random array and applied to practical indoor and outdoor sources. The experimental results reveal that the optimized random array is capable of imaging large-scale noise sources situated in long distance.