Research On Key Techniques Of Passive Sonar Target Detection In Shallow Sea Area

In shallow water area,passive sonar shoulders the task of large-scale sea area monitoring and port warning,and often works under the condition of multi-target,multi-interference and low signal-to-noise ratio.Therefore,target detection often faces three problems: when there are multiple targets,the conventional processing algorithm has insufficient detection and resolution for weak targets;The detection of submerged weak targets is seriously affected by the large amount and intensity of sea surface interference;There are many targets detected,which makes it more difficult to distinguish between submerged and surface targets.Multi-target detection and resolution,strong interference suppression,and the resolution of submerged and surface targets are the key technologies affecting passive sonar performance in shallow water.Starting from the phase structure of the received signal,this paper studies the above three technical problems by using the spatial phase structure,the frequency phase structure,the modal phase structure and the wavenumber domain phase structure respectively,and establishes the relevant algorithm model.In view of various complex conditions and model parameters,on the basis of theoretical analysis,this paper presents a new method to solve these problems.A large number of simulation data and experimental data are used to analyze and verify the algorithm,and good results are achieved.The research in this thesis has important significance and engineering application value for improving the detection performance of passive sonar target.Aiming at the problem of weak target detection and azimuth resolution under multi-target condition,this paper proposes solutions from several aspects.Firstly,aiming at the shortcomings of the existing inverse beamforming algorithm based on acoustic field integral,an improved algorithm is proposed by using a new covariance matrix Toeplitz averaging process from the perspective of spatial phase structure.The simulation and experimental data validation show that the new algorithm can improve the azimuth resolution of the target,while ensuring the detection ability of weak targets,which is better than the performance of the existing inverse beamforming methods.Secondly,a passive sonar target azimuth estimation algorithm based on frequency diversity technology is proposed from the perspective of frequency domain phase structure.The signal processing gain and multi-target azimuth resolution are improved by using the received signal frequency domain phase structure and phase compensation or reconstruction between frequencies.According to the different application background of two hydrophones,linear array and co-prime array,the deduction process of the algorithm principle and the theoretical analysis of the algorithm performance are given respectively.The effectiveness and robustness of the proposed algorithm are proved by a large number of simulation and experimental data.Aiming at the problem of water surface interference affecting submerged weak target detection,based on the theory of shallow water waveguide,the dispersion characteristics and the variation of modal intensity with depth and propagation distance during normal mode propagation are analyzed,and the warping transform and modal filtering method utilizing dispersion characteristics are given.On this basis,three methods of surface interference suppression based on modal distribution for vertical array are studied.Because the application of vertical array is limited,a surface interference suppression algorithm for horizontal array is proposed.Based on the difference between horizontal linear array and horizontal circular array in modal decomposition technology,a method of interference suppression based on horizontal circular array is proposed.According to the modal domain beamforming algorithm,the azimuth estimation results of each order mode are calculated,and the conventional azimuth estimation is performed by using the modal distribution characteristics and the multi-order modal estimation results.The results are revised to restrain the disturbance of water surface.The simulation and experimental data validation show that the mode intensity extraction and interference suppression based on circular array are better than that based on horizontal array.Aiming at the problem of discrimination between surface and submerged target,two methods of discrimination for horizontal array are proposed in this thesis from the point of underwater acoustic physics(modal characteristics).The first method is based on the phase structure in the modal domain of the horizontal linear array.The modal domain beamforming is used to estimate the modal intensity of each order,and the matched modal processing is used to estimate the depth of the target.Although this algorithm can give the estimation result of target depth,it needs many conditions,such as multi-mode number,large aperture,ocean environment parameters and target azimuth.The second method calculates the energy distribution of signal in f-k domain based on phase structure in wavenumber domain,and distinguishes surface and submerged targets according to their distribution characteristics.The principle of using this method is that the energy distribution of targets with different depths is different in modal domain.Although this method can not give accurate target depth,compared with the existing depth identification algorithm based on shallow-water waveguide theory,it can better distinguish water surface targets and submerged targets below a certain depth under the condition of small aperture of horizontal array and less information of marine environment,and it is more suitable for practical application background.Moreover,it is not affected by the moving condition of the target(causing signal changes),so it has robustness in physical mechanism.