Research on Port-Starboard Discrimination Performance of Triplet Arrays under Different Noise Conditions

Abstract

The triplet arrays provide the ability to solve the notorious port-starboard (PS) ambiguity problem of ordinary single line array. Cardioid beam pattern is provided by the triplet arrays structure by using the time delay of received signals and noise correlation model. The received noise is composed by ambient sea noise and self noise. The former is made of wind and shipping noise, the latter is made of towed flow noise and electric noise. The PS performance is worse when the flow noise increases. The correlation coefficient of triplet arrays is researched by simulation under different noise conditions. The minimum signal to noise ratio (SNR) of PS discrimination is analyzed for cardioid beamformer and adaptive beamformer. The ability to PS discrimination and anti-noise/jammer of adaptive beamformer is better than cardioid beamformer by processing simulation and trial data. Introduction The underwater target is detected, tracked, localized and classified through acoustic array and array signal process by sonar devices. The PS ambiguity problem can’t be solved by ordinary single line array, because one hydrophone is omnidirectional. At the beginning of the 90’s, the triplet arrays are firstly introduced to solve PS ambiguity by Jean Bertheas. The method using noise correlation of triplet arrays is given by Y Doisy to solve PS ambiguity.The three hydrophones of one triplet are treated as one directional hydrophones. Cardioid beampattern is provided by the triplet arrays structure by using the time delay of received signals and noise correlation model [2-5] at the infinite PS rejection principle. Thus, the PS sides of the target can be discriminated. The PS discrimination performance of cardioid beamformer is optimum when the received noise is highly correlated. The PS discrimination and array gain is loss when the flow noise increases. The adaptive beamformer of triplet arrays is a beam-domain beamformer. The weights are computed using the on-line estimated correlation matrix. If the correlation matrix is properly estimated, the ability of PS discrimination and array gain are optimum, and the sidelobe of PS beamformer is lower. In the paper, the noise correlation coefficient under different noise conditions is researched by simulation, and meanwhile the PS discrimination performance is given for cardioid beamformer and adaptive beamformer. Principle of port-starboard discrimination of triplet arrays The layout-structured graphic of hydrophone triplet is shown in Fig.1, where θ is the angle between incoming signal and y axis, and φ is the angle between projection of incoming signal in xoz plane and z axis. Position of hydrophone triplet in the vertical plane is: ) 3 / 2 , 3 / 2 , 0 ( ) , , ( 3 2 1 π π φ φ φ − = . International Industrial Informatics and Computer Engineering Conference (IIICEC 2015) © 2015. The authors Published by Atlantis Press 1666 Fig.1 Inner structure of hydrophone triplet The received signals, which contain signals from targets and complex noise, can be divided into two classes: non-correlation and correlation noise. The former is mainly flow noise and electrical noise, and the latter mainly includes ambient sea noise, shipping noise and traffic noise. The background noise of the triplet arrays is regarded as directive noise with the same correlation property. Thus, the noise correlation matrix of triplet arrays can be decomposed into two parts: