Orthogonal linear array configurations for underwater 3D acoustical imaging: Performance study and a nonlinear beamforming approach

While underwater acoustical three-dimensional (3D) imaging offers several benefits compared to two-dimensional imaging, the requirement of a uniform planar array (UPA) with a large number of elements for image reconstruction limits its widespread use. The most popular array structure capable of estimating the direction of arrival in 3D space with low complexity is a cross-array, where two linear arrays are placed orthogonally in a symmetric arrangement. Although other orthogonal linear array configurations are feasible, they have not been thoroughly investigated in the context of underwater 3D acoustical imaging. In this work, we studied six different orthogonal array configurations for underwater acoustical 3D imaging. The study proved that an L shaped array placed at the edges of a UPA performs better than any other orthogonal array both in terms of resolution and ability to resolve ambiguity in multiple targets. We also propose a nonlinear beamforming approach using an L-shaped array for real-time 3D image reconstruction. The proposed approach achieves a signal-to-noise ratio of 12.72 dB and a contrast-to-noise ratio of 1.35 while reducing computation time by 94.64% through the use of significantly fewer array elements than conventional delay-and-sum beamforming using a UPA.