Acoustic Lucky Imaging for microphone phased arrays

Abstract

Acoustic imaging methods are used to detect and quantify aerodynamic noise sources in aeroacoustic experiments. In wind tunnels with an open jet test section the sound wave propagating from an aeroacoustic source to a microphone array is distorted by the shear layer in the propagation path. The velocity fluctuations caused by the turbulence in the shear layer continuously alter the propagation time between source and observer causing reduction of the coherence between two microphone signals. The reduction in coherence is stochastic and dependents on the relative position between source and microphone. This leads to blurred acoustic source images with reduced peak Sound Pressure Level. This paper presents the Acoustic Lucky Imaging methodology, based on a methodology in astronomy to correct for image distortion due to turbulence present in Earth’s atmosphere. Furthermore, a model for coherence loss is derived to explain the loss of acoustic image resolution and allow for a-priori estimates on acoustic image blurring. The methodology is applied to an acoustic data set obtained in the open jet of a large industrial wind tunnel. It is shown that the presented methodology can restore a loss of 6 dB peak Sound Pressure Level by 3 dB and increase the resolution of the acoustic image at 8 kHz, a wind tunnel speed of 34 m s−1, and a shear layer thickness of ~1 m.