Research On The Key Techniques About Direction Of Arrival Estimation Based On Vector Hydrophone Array

In recent years,the competition between our country and neighboring countries in territorial water sovereignty and marine resource development has become increasingly prominent.With the help of new materials of anechoic tiles and shock absorption technology,countries all over the world vigorously develop stealth and quiet submarines and other weapons and equipment.As a result,the frequency and intensity of new equipment radiating noise are significantly reduced,which makes it more difficult for traditional acoustic pressure hydrophone array to detect targets in low frequency band.How to improve the detection performance of underwater low frequency target is an urgent problem.As a new type of underwater acoustic sensor,the acoustic vector hydrophone,which has frequency independent dipole directivity,has opened up a new field of vision for passive detection.It is able to simultaneously measure the pressure and particle velocity components in space.By making full use of available acoustic information,the detection ability of low frequency targets could be effectively improved.Therefore,it has attracted wide attention from scholars at home and abroad.Combined with the complex underwater environment,this thesis aims to improve the estimation performance of low frequency targets in the presence of limited array aperture.The researches of the thesis include the direction of arrival(DOA)estimation method under small aperture vector hydrophone array,the off-grid DOA estimation method,and the robust DOA estimation methods in the presence of axial inconsistent and non-orthogonal factors via the acoustic vector hydrophone array.A series of theoretical analysis,computer simulations and experimental verifications are carried out to address these themes,which lays the foundation for engineering application.The main research results and innovations are summarized as follows:1.To address the problem that the traditional high-resolution algorithms and the existing sparse signal recovery algorithms degrade the resolution performance of the small aperture vector hydrophone array,an iterative sparse covariance matrix fitting algorithm(ISCMFA)is proposed.Firstly,the signal covariance matrix is obtained by eigenvalue decomposition.Then,based on the weighted covariance matrix fitting criterion,the objective function via signal covariance matrix and sparse signal power is constructed,and the analytical expression of sparse signal power is derived by using the properties of Frobenius norm.The result shows that the signal power obtained by the ISCMFA method is preprocessed by the filter,which allows the signals in the specified direction to pass through and attenuate the signals in other directions,thus reducing the sensitivity to coherent sources,and improving the resolution performance of coherent signals.On this basis,a sparse signal power iterative compensation(SSPIC)method is proposed.The resolution performance of small aperture vector hydrophone array is further improved by compensating the signal power on all grid points.The simulation results show that the proposed methods improve the resolution performance of the small aperture vector hydrophone array without losing the degree of freedom of it.2.In order to solve the off-grid issue in sparsity-based DOA estimation for vector hydrophone array,an off-grid alternating iterative weighted(OGAIW)method is proposed.In the proposed method,the off-grid deviation between the real signal azimuth and its nearest grid position is introduced into the signal model as an unknown disturbance parameter,and the reconstructed interference plus noise covariance matrix is used as the weighting term.Then,the cost function via sparse signal and unknown disturbance parameter is formulated.For fixed off-grid disturbance parameter,the signal corresponding to coarse grid position is estimated by weighted least square method;for fixed sparse signal,the off-grid disturbance parameter is acquired by weighted least square method,and the sparse signal and off-grid deviation parameter are updated alternately.In the case of convergence,the estimated off-grid deviation parameter is used to modify the estimated azimuth under the coarse grid as the final estimation of the target azimuth.Simulation results show that,compared with the existing algorithms,the proposed method improves the DOA estimation accuracy in the case of coarse sampling grid,and has faster convergence speed.3.The DOA estimation performance for a uniform vector hydrophone linear array with axial inconsistency is studied.The signal model via vector hydrophone array in the presence of axial inconsistency is introduced firstly.Then,the influence of axial deviation on beam directivity of vector hydrophone array is analyzed.The analysis results show that the axial deviation causes the beam deviation of vector hydrophone array.Moreover,the closed solution of DOA estimation error is derived by using the first order Taylor series expansion.The results show that the DOA estimation error is related to the number of the sensor,the DOA of the source and the axial deviation.In order to improve the accuracy of DOA estimation under axial bias,an alternating iteration adaptive approach(AIAA)is proposed.In the proposed method,the axial angle deviation parameter is introduced into the signal model,and then the sparse signal and axial deviation matrix are estimated by alternating iteration.In each iteration,in order to eliminate the influence of axial angle deviation on DOA estimation,the axial deviation matrix reconstruction method is proposed based on the inherent directionality of vector hydrophone.Simulation results verify the effectiveness and accuracy of AIAA,and prove that the proposed axial deviation matrix reconstruction method can improve the DOA estimation performance of the axial inconsistent vector hydrophone array by combining the existing DOA estimation methods.4.Aiming at the problem that the DOA estimation performance of existing algorithms using the non-orthogonal hydrophone is degraded,a non-orthogonal deviation matrix reconstruction(DMR)method is proposed.Firstly,the DOA estimation performance of two non-orthogonal vector hydrophone models is studied,and the DOA estimation bias of them is quantified.The theoretical analysis results show that the non-orthogonal parameters will cause the estimated azimuth to deviate from the true direction of the source.In order to improve the accuracy of DOA estimation of the non-orthogonal vector hydrophone,a reconstruction method of deviation matrix is proposed based on matrix rotation technique.The results show that the method combined with traditional DOA estimation methods can improve the DOA estimation accuracy of the non-orthogonal vector hydrophone.The DOA estimation performance of non-orthogonal vector hydrophone array is also studied.Two non-orthogonal vector hydrophone array models are established,and the analytical expression of DOA estimation error caused by non-orthogonal bias is derived.The results show that the non-orthogonal bias has a greater impact on the performance of the vector hydrophone array when the velocity hydrophone in x-axis is placed correctly than that in y-axis.In order to improve the accuracy of DOA estimation for non-orthogonal vector hydrophone array,the ISML-MAP method is proposed.In each iteration,three matrix rotation methods are introduced to estimate the non-orthogonal deviation matrix.Finally,the DOA estimation is realized by searching the spectrum peak of sparse signal power.Simulation results demonstrate that the proposed method achieves better DOA estimation performance than the conventional techniques over a non-orthogonal AVS array.5.Based on the theoretical analysis and computer simulation,the water tank experiment system of single vector hydrophone and the lake exprement system of vector hydrophone array are formulated.The proposed methods in Chapter 3,Chapter 4,Chapter 5 and Chapter 6 are carried out in tank and lake experiments.The experimental results show that the ISCMFA and SSPIC methods in Chapter 3 have higher resolution than the existing methods when the signalto-noise ratio(SNR)is low;the CPU running time of the OGAIW method in Chapter 4 is close to that of the classical method in coarse grid,and the performance is better than the classical methods and the existing off grid methods;In Chapter 5,the reconstruction method of the axial deviation matrix combined with the existing DOA estimation methods improves the accuracy of DOA estimation of the axial inconsistent vector hydrophone array;the proposed DMR method in the Chapter 6 can combine with existing DOA estimation methods to improve the DOA estimation accuracy under two non-orthogonal models for a single vector hydrophone.The results of tank experiment and lake expriment verify the effectiveness and feasibility of the proposed methods in this paper,which lays a foundation for the practical engineering applications.