Research On Key Technologies Of Passive Location For Underwater Target On Moving Platform

The positioning of passive targets underwater is a very important guide for both security and attacking threatening enemy targets.Position information is a stable feature of underwater targets and passive positioning based on observed position is a proven means of locating them.Passive sonar for target detection on mobile platforms often faces special problems in its application.Passive sonar on mobile platforms can be affected by non-ideal factors such as restricted aperture of the base array,limited system energy,near-field interference and platform motion,which pose challenges to sonar signal processing.Existing algorithms are unable to meet the demands of these non-ideal environments on array performance.In order to meet the demand for high-resolution direction of arrival(DOA)estimation and localisation of mobile platform mounted sonar arrays in non-ideal environments,this paper focuses on sonar array design,direction-of-arrival estimation under time-varying observation array positions,direction of arrival estimation under near-field spatial colour noise,and key technologies for passive localisation based on the observed direction-of-arrival estimation of mobile platforms,and conducts in-depth analysis and research,aiming to lay the foundation for passive localisation of mobile platforms.The main research of this paper are as follows.For the passive sonar of target detection carried by the mobile platform,in order to ensure that the visible area does not show the fence phenomenon,it is necessary to ensure that the interval between adjacent array elements of the uniform line array is not greater than half of the wavelength.The relative bandwidth of passive sonar detection is wide,and for higher frequency components,the use of uniformly spaced arrays requires more transducer elements,making the array system too complex and increasing the processing power requirements.To address this problem,this chapter combines the characteristics of vector hydrophones themselves and proposes a compressive sensing based array design technique for vector arrays.The technique combines the characteristics of vector hydrophone arrays to reformulate the compressive sensing problem.By transforming the weights in the objective function and constraints,the optimally designed array can obtain fewer array elements while satisfying the beam map performance requirements.By introducing the de-gridding method,the optimised beammap can be better matched with the ideal beammap.Finally,the simulation experiments verify the superiority of the algorithm.In order to address the problem that the relative direction of arrival estimation of the target and the base array under the random rotation scenario of the platform varies rapidly in time,it is difficult for the subspace-like direction of arrival estimation method to obtain a large number of effective snapshots of the array data,which in turn leads to the degradation of the direction of arrival estimation estimation performance for weak targets in the far field,a direction of arrival estimation estimation technique based on space domain rotation focusing is proposed.Firstly,changing heading angle,an array reception model is constructed for the platform motion.Based on this,the a priori information of the real-time heading angle output from the platform’s magnetic compass is used to design an airspace rotation matrix,which is able to compensate for the relative change of target direction of arrival estimation information in the data snapshots due to the change of heading angle by rotation,so that the target direction-of-arrival estimation information can be focused in the airspace,and thus the covariance matrix can be estimated accurately using a large number of continuous data snapshots.The covariance matrix is applied to a subspace-like direction of arrival estimation method to achieve high-resolution direction of arrival estimation estimation of weak targets in far-field in a platform rotation scenario.Compare the performance of the algorithm in terms of direction of arrival estimation accuracy and direction of arrival resolution.The direction of arrival method based on sparse Bayesian learning in a colour-noise background is proposed to address the problem that the background noise does not satisfy the assumption of isotropy due to the interference of the near-field volume source of the platform,which in turn affects the direction of arrival estimation performance of the high-resolution method estimation algorithm for far-field targets.The proposed method makes full use of the advantages of the sparse reconstruction class algorithm,but also refines the spatial colour noise model,and improves the estimation accuracy of the far-field target direction of arrival in spatial directional noise fields by jointly and iteratively solving the signal power and noise parameters for each angle in the air domain under the framework of sparse Bayesian learning.The results show that the proposed method does not show any significant increase in the pseudo-peaks and noise spectral levels while ensuring the high resolution of the sparse reconstruction-like algorithm in the colour-noise background,and the proposed algorithm can obtain higher resolution,estimation accuracy compared with conventional high-resolution direction of arrival methods.Finally,the effects of the canonical parameters and the order of the linear noise model on the performance of the algorithm are discussed through computer simulations,and the performance of this direction of arrival estimation algorithm is verified.In localisation of underwater targets by direction of arrival estimation using motion arrays,Newton’s algorithm is used to solve the non-linear optimisation problem in traditional underwater localisation based on direction-of-arrival measurements.However,in real scenarios,where accurate initial values are generally not available and there are situations where the initial values are far from the actual values,the convergence probability of Newton’s algorithm is reduced,which can significantly reduce the positioning accuracy.At the same time,different moments of the motion array can be used as localisation nodes,which bring about a large number of localisation nodes,and it is a challenge to optimise the localisation nodes in order to achieve real-time processing.In order to solve the above problems,this chapter proposes an underwater localisation method based on an improved Newtonian algorithm.The proposed algorithm firstly proposes an improved Newton’s method by introducing a singular value factor,which corrects the pathological partial derivative matrix and improves the convergence probability of the algorithm.Secondly,a localisation node optimisation algorithm is proposed to achieve efficient selection of a large number of localisation nodes.The algorithm optimises the localisation accuracy and meets the real-time requirements.Finally,simulation is carried out in this chapter,and the results match with the theoretical analysis,proving the effectiveness of the method.