The Effects of Vertical Correlation Characteristics on Vertical Array Gain Performance in Deep Water

• Published in: IEEE journal of oceanic engineering Vol. 47; no. 3; pp. 751 - 766
• Main Authors: Dong, Fanchen, Li, Zhenglin, Hu, Zhiguo, Wu, Shuanglin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.07.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Array signal processing; Arrays; Bathymetry; Beamforming; Coefficients; Correlation; Correlation coefficient; Correlation coefficients; Covariance matrices; Covariance matrix; Deep water; Delay effects; Detection; Eigenvalues; Eigenvalues and eigenfunctions; Environmental conditions; Signal to noise ratio; Sonar arrays; Sound field in deep water; Sound fields; Transmission loss; uneven bottom; vertical array gain performance; vertical correlations
• ISSN: 0364-9059; 1558-1691
• DOI: 10.1109/JOE.2021.3125970

Spatial correlation characteristics of sound field have significant impact on array gain. Compared with the single-sensor receiving scenario, the vertical array with multiple receivers can remarkably enhance signal-to-noise ratio and detection ability. For the large-scale vertical array deployed in the incomplete deep-water sound channel, different bathymetry conditions or source-receiver ranges lead to different vertical correlation patterns, thus leading to the variation of array gain. Based on the deep-water experimental data conducted in the South China Sea, the effects of the vertical correlation on array gain performance are studied in both the flat and the uneven bottom environments. In the flat bottom environment, three representative ranges in the direct arrival zone and first shadow zone are chosen, respectively, to analyze how vertical correlation characteristics would affect the array gain performance. In the uneven bottom environment, attention is confined to those ranges that endure significant transmission loss differences when compared with the flat bottom environment. In summary, when the correlation coefficients grow smaller, it is observed that the array gains of different beamformers decrease, along with the energy drifting away from the first few eigenvalues and dispersing on many other eigenvalue elements. At the same time, the gap between the array gain results of eigenvalue beamforming and optimal quadratic beamforming would grow wider and the effective rank of signal covariance matrix would sprawl as well. The study is of great significance to analyze and enhance vertical sonar array detection performance at different ranges under various environmental conditions in deep water.