Research On Depth Discrimination Technology Based On Combined Hydrophone

It is very important to discriminate whether the target is surface or submerged for protecting itself and attacking the enemy's threatened targets.In the shallow water waveguide,the performance of passive sonar mounted on mobile platform is not only limited by the complexity of shallow water wavegudie,but also affected by the limited aperture of array,lacking of prior information of target and mismatched environment,which bring challenges to source discrimination.Considering the practical difficulties of passive sonar on mobile platform,according to the connection and difference of acoustic field characteristics of pressure field and vibration velocity field,this paper focuses on some key technologies of passive sonar in source discrimination.The main research contents include:When the aperture of horizontal line array(HLA)is limited,it will affact the extraction of classification feature quantity and the performance of the source discriminator.In order to solve the problem,we propose two methods based on modal features in this paper.Firstly,we analyze the amplitude distribution of modes excited by surface and submerged.The analysis results show that when the receiver is far away from surface,amplitude distribution of modes excited by surface and submerged is different.On this basis,the normalized low-order modal power is used to describe the amplitude distribution characteristics of modes,and it is used as the classification feature.In order to extract the classification feature,a joint modal beamforming(JMBF)algorithm is proposed in this paper.The algorithm makes use of the fact that the sound pressure and vibration velocity are not related in the noise field,and improves the extraction effect of the classification feature from the perspective of improving the processing gain.According to the estimation results of the classification feature and the related concepts in the detector,a JMBF discriminator is designed.In addition,the normalized low order modal strength is defined to describe the amplitude distribution characteristics of modes and is taken as the classification feature.In order to extract the feature,we propose a feature estimation algorithm based on reduced rank total least square(RRTLS).The algorithm improves the estimation performance of classification feature quantity by reducing the influence of interference normal mode on the feature estimation performance.According to the estimation results of the classification feature,a RRTLS discriminator is designed.In this paper,two methods are proposed to improve the performance of feature estimation from the perspective of processing gain and interference normal mode suppression,and then improve the classification ability of source discrimination under the condition of array aperture limitation.Finally,the effectiveness of the proposed methods are verified by simulation experiments.When estimating the depth of a moving target using a single hydrophone,the resolution of the wavenumber spectrum will be insufficient due to the small moving range,which will affect the performance of source depth estimation.Firstly,source depth estimation method using a single scalar hydrophone is studied.The technology makes use of the characteristics of high resolution of auto regressive model(AR),which can improve the estimation effect of horizontal wavenumber in the case of unknown source range and the number of horizontal wavenumbers,and then effectively improve the depth estimation ability under the condition of small moving range.Then,the method is extended to single velocity hydrophone,and source depth estimation using a single velocity hydrophone is studied.The combined hydrophone is composed of a pressure hydrophone and a velocity hydrophone.The sound pressure and vibration velocity carry the source depth information in different ways.The joint processing of sound pressure and vibration velocity can further improve the depth estimation ability using combined hydrophone.Four kinds of combination methods of sound pressure and velocity wavenumber spectrum and the cost function of depth estimation are discussed using a single combined hydrophone,and the applicability of each combination methods is studied.In this paper,the source depth estimation method proposed in this paper can improve the source depth estimation ability using single scalar hydrophone by improving the resolution of wavenumber spectrum.In addition to the high resolution of AR wavenumber spectrum,it also further improves the target depth by jointly processing the spectrum of sound pressure and vibration velocity using a single combined hydrophone.Finally,the effectiveness of the proposed method is verified by simulation experiments.Target distance is an important auxiliary information of source discrimination,and obtaining this information is helpful to improve the performance of source discrimination method.In this paper,a passive range estimation method based on the characteristics of continuous spectrum of target is studied to solve the problem of source range in quasi pekeris waveguide and strong negative gradient waveguide using a single combinated hydrophone.According to the dispersion characteristics of reflected and refracted modes,the warping operator expressions of the cross-correlation functions of sound pressure and vibration velocity in quasi pekeris waveguide and strong negative gradient waveguide are derived respectively.The physical characteristics of the frequency spectrum obtained from the warping transformation of the cross-correlation function of sound pressure and vibration velocity are analyzed.The analysis results show that the peak frequencies of the spectrum spectrum carry the target distance information.Based on the physical characteristics of frequency spectrum,three passive ranging methods are proposed according to different waveguide types and the existence of guided sound source: passive ranging method based on guided sound source spectrum information,source range estimation method based on matched spectrum characteristics and source range estimation method under strong negative gradient waveguide environment.The source range estimation method based on guided source spectrum information and matched spectrum feature can be used to solve the problem of source range estimation in quasi pekeris waveguide environment.Among them,the source range estimation method based on guided source spectrum information requires a guide sound source with known range in the waveguide,and the source range estimation method based on matched spectrum feature requires relatively accurate prior environmental information.Source range estimation method method in strong negative gradient waveguide environment can be used to solve the problem of source range estimation in strong negative gradient environment,which requires a guide sound source with known distance in the waveguide.In practical application,the appropriate algorithm can be selected according to the type of actual waveguide,the existence of guide source and the accuracy of prior environmental information.Finally,the effectiveness of the proposed technology is verified by simulation and sea trial.The research is based on the characteristic of channel.And the research results of this paper can be applied to source depth discrimination and source range estimation.