A monotonic two-step iterative shrinkage/thresholding algorithm for sound source identification based on equivalent source method

Abstract

Near-field acoustical holography (NAH) based on equivalent source method (ESM) is an efficient technique applied for sound source identification. However, the conventional ESM with ℓ2 norm regularization cannot produce a satisfactory solution in high frequency, where the average array inter-element spacing is larger than half a wavelength. Therefore, conventional ESM with Tikhonov regularization is restricted to relatively low frequency. To overcome the issue, an alternative method called monotonic two-step iterative shrinkage/thresholding algorithm for near-field acoustical holography is proposed. In the algorithm, another existing algorithm called Wideband Acoustical Holography (WBH) is used to generate the threshold, and also be used as a benchmark for comparison. Simulated measurements based on an irregular microphone array are performed to compare the performances of the Tikhonov regularization, WBH and MTwIST. The results suggested that the proposed method can identify the sound sources more accurately than Tikhonov regularization in full frequency range, and it performs better than WBH in the case of identifying coherent sources at relatively low frequency. The experiments demonstrated the validity and the practicability of the proposed method.