Application of Inverse Patch Transfer Functions Method With Wideband Holography Algorithm to Sparsely Distributed Sources Identification

Abstract

Inverse patch transfer functions (iPTF) method has been developed to reconstruct the sound field of irregularly shaped sources in a noisy environment. The iPTF method, which uses classic regularization methods to solve the ill-posed problems generally, would incur some sidelobes ghosting in the process of identifying sparse sources. In view of the fact that the algorithm in wideband holography (WBH) can promote sparsity of results, a technique combining iPTF method with WBH algorithm is proposed to identify sparsely distributed sources in the present work. In the proposed technique, double layer pressure measurements are used to replace the measurements of the pressure and normal velocity which uses costly p-u probes. A gradient descent algorithm and a filtering process are applied to solve the minimization problem of identifying the normal velocities of target sources, which can suppress ghosting sources rapidly by an iterative process. In simulations, the field reconstruction results of two antiphase square piston sources show good sparsity and accuracy by employing the technique, nearly without ghosting sources. At different distances and frequencies of the two sources, the technique still performs well. Experimental validations at 200 Hz and 400 Hz are carried out in the end. The results of experiments are also coinciding with those of simulations.