Source localization using beamforming and double L-shape sensors arrays

Beamforming or phased array is widely used in structural health monitoring (SHM) systems for source localization. This localization is achieved by the constructive interference of signals measured by sensors in different positions. The uniform linear array (ULA) is one of the most commonly used beamforming array layouts due to its low complexity and high reliability. Unfortunately, ULA has the disadvantage of very low resolution when the direction of arrival (DOA) approaches certain directions. In order to overcome this problem, the double L-shape array is employed by adding only one sensor to the ULA. This sensor array uses the time difference of arrival information to estimate the DOA, which has high resolution in all directions except for the direction very close to vertical. Due to the blind area of the ULA beamforming and double L-shape sensors array, we combine the two methods and proposes a novel combined sensor array configuration which has no blind area at all. In addition, this novel combined sensor array layout uses a fewer number of sensors, which can reduce the cost of SHM practice significantly. Several numerical investigations are carried out to demonstrate the performance of the proposed sensor array.

I. Introduction

Structural health monitoring (SHM) has been one of the many applications of beam-forming technology. Giurgiutiu et al. [1] used beamforming algorithm and embedded-ultrasonic structural radar (EUSR) to monitor a thin-wall structure. Yu et al. [2] utilized the linear EUSR phased array beamforming detect cracks successfully in 2007. In 2008, Yu et al. [3] continued the study of in situ beamforming method for damage detection in thin-wall structures using 64 embedded sensors. Although the detection accuracy has been improved, a large number of sensors are used, so it is not suitable for practical engineering applications. In 2010, De Marchi et al. [4] presented a warped frequency transform and delay-and-sum beamforming-based procedure to process dispersive Lamb waves for SHM applications. From 2011 to 2014, He et al. [5]-[7] used near-field beamforming method to localize the rotor-stator rubbing fault in rotating machinery. In 2018, Jung et al. [8] proposed an asymmetric sensor array to reduce the ‘spatial aliasing’ effects in beamforming-based source localization applications. However, the inherent disadvantage of the beamforming method-the existence of blind spots at certain angles-has not been resolved.