浅海混响环境中主动自导的空时自适应处理技术研究

发布时间：2018-03-10 12:54

本文选题：空时自适应处理　切入点：任意形状海底　出处：《西北工业大学》2014年博士论文　论文类型：学位论文

【摘要】：对于浅海混响下主动自导系统来讲，混响主要来自海底和海面。混响作为浅海自导系统特有的干扰，对水声信号检测性能的影响日趋凸显，由于混响和目标回波具有很强的相关性，对混响抑制一直是水声信号处理研究中的热点和难点。本文研究了浅海混响中主动自导系统空时自适应处理技术，该研究属于水声信号处理领域的学术前沿，具有重要的理论意义和应用价值。主动自导系统一个重要的特征是其运动引起的混响多普勒扩展，且这种扩展与雷达的杂波谱扩展特性非常相似，空时自适应处理方法抑制雷达杂波的效果理想，因此，对浅海环境中自导系统混响的抑制可以借鉴这样成熟的杂波抑制技术。针对自导系统在浅海环境下的应用条件，采用空时自适应算法对来自任意海底和海面的混响进行抑制。本文主要研究成果与创新点如下： 1、建立了任意形状海底的混响谱模型。针对海底凹凸不平的实际情况，用曲面方程描述实际海底形状，通过牛顿迭代法求出海底任意点的坐标，根据海底单元与主动自导系统的相对位置关系和方位信息计算不同多普勒单元的混响大小与发射信号能量谱卷积得到任意海底混响谱，解决了现有的模型将海底统一看作平面建模带来的不足，进一步推导给出了谱模型的简化计算方法。从混响衰减曲线、混响级、概率分布以及频域和空间相关性多个方面对混响特性进行了分析，从仿真分析和误差对比验证了谱模型的正确性。 2、建立了海面混响模型。研究了海面混响特性及形成机理，考虑到海面的动态变化即谱的分布，，优化提出了动态海面分形模型。推导并给出了海面散射强度的表达式以及在不同海面因素作用下具体的海面散射强度计算公式。通过仿真分析验证了各因素对散射强度的影响及变化规律。 3、研究了主动自导系统的混响特性。推导了主动自导系统浅海混响的空时分布，仿真验证了自导系统在不同安装角度下混响分布特性。通过对STAP混响抑制方法及空时最优处理性能的研究，得到可通过低阶的空间近似全维数据空间或者向低维的子空间投影，实现降维处理。针对空时最优处理中协方差矩阵估计计算量大、难以实现等问题，对主动自导系统的混响自由度进行了分析，同时给出了混响子空间维数的计算方法，为空时自适应降维研究提供基础。 4、研究了主动自导系统降维STAP混响抑制。深入研究了TSA、M-CAP空时自适应降维算法，从改善因子、运算量、数据采样、实现复杂度等各方面分析了降维算法与最优处理相比的优劣性。仿真给出了不同降维算法的空时处理结果及改善因子曲线。为研究新的STAP降维混响抑制方法奠定基础。 5、提出了基于信息熵的STAP混响抑制方法。在对空时降维算法的研究基础上，引入了信息熵概念做特征数据压缩，根据特征值分解后目标与混响所表现出的特性不同，对采集的空时二维数据进行数据选择，通过仿真对算法进行分析验证。 6、提出了基于信息熵的降维STAP混响抑制方法。针对基于信息熵的STAP混响抑制方法中特征值分解带来的算法复杂度增加问题对算法进行了改进。在信息熵数据提取的基础上，对接收的数据进行子数组划分，然后做空时平滑处理来估计协方差矩阵，原始的空时处理器维数由NK变为PQ，其中（P N, QK），根据最小线性约束准则求得自适应权向量，对目标检测数据采用大小为PQ的空时窗用权矢量平滑加权输出。 7、对基于信息熵的降维STAP混响抑制方法进行了实测数据的实验验证。从原始数据的采集到最后的输出处理作了详细的分析，实验分析结果表明该算法可以在降低计算量的同时较好的检测出目标。 8、针对本文所建的任意形状海底混响模型，采用基于信息熵的降维STAP算法进行任意海底混响抑制。用曲面方程对海底平面进行模拟，通过牛顿迭代算法得到曲面方程解计算不同位置的方位和俯仰角。根据仿真得到的混响空时分布结果，采用基于信息熵的降维STAP方法，计算混响协方差矩阵和权矢量并总结了海底混响抑制的步骤。仿真分析验证了任意海底形状在不同的曲面方程下混响级、多普勒频率的变化规律及空时输出结果。 9、针对本文所建的海面混响模型，采用基于信息熵的降维STAP算法进行海面混响抑制。在动态海面模型基础上，分析了不同因素对海面混响的影响及不同因素作用下自导系统的海面混响空时谱和功率谱分布特性。根据仿真得到不同因素作用下的混响空时分布，将海面划分为粗糙面作用与含气泡粗糙面作用两种情况，给出了两种方式下的海面空时处理结果。
[Abstract]:For under the shallow water reverberation active homing system, mainly from the seabed and surface reverberation. The reverberation in shallow water as homing interference system peculiar to the influence on the acoustic signal detection performance is becoming increasingly prominent, because the reverberation and target echo has a strong correlation, the reverberation suppression is a hot and difficult research of underwater acoustic signal processing in this paper. The active homing system of space-time adaptive processing technology in the study of shallow water reverberation, which belongs to the academic frontier in the field of underwater signal processing, and has important theoretical significance and application value.
One of the important features of the active homing system is its motion caused by reverberation Doppler expansion, and the expansion of clutter and radar propagation is very similar to that of the ideal, the effect of space time adaptive processing for clutter suppression and clutter suppression technology that can learn from the mature inhibition of homing system reverberation in shallow water. According to the application conditions in shallow water environment system, the space-time adaptive algorithm of arbitrary seabed and surface reverberation suppression. The main research achievements and innovations are as follows:
1, any shape of the sea floor reverberation spectrum model is established. According to the actual situation of the uneven seabed, seabed shape with a description of the actual surface equation, using the Newton iteration method is used to calculate the coordinates of any point of the seabed, according to the unit and the active homing system of the relative position and position information to calculate the different Doppler unit size and reverberation signal the energy spectrum obtained by convolution with arbitrary reverberation spectrum, the solution of the existing model will be unified as lack of submarine plane modeling brings, further calculation method was deduced. The simplified spectrum model of reverberation attenuation curve, a reverberation level, probability distribution, frequency and spatial correlation analysis of reverberation characteristics, correct spectrum the model was verified through simulation analysis and error analysis.
2, establish the model of sea surface reverberation. The reverberation characteristics and formation mechanism, considering the dynamic changes of the sea spectrum distribution, optimize the proposed dynamic sea fractal model. The expression is given the scattering intensity as well as in different sea factors specific sea surface scattering strength calculation formula is verified by simulation. The influence of scattering intensity and the variation of various factors.
3, study the characteristics of reverberation active homing system. Deduced the active homing system of the reverberation time distribution, simulation of the distribution of the reverberation homing system in different installation angle characteristics. By inhibiting the study of optimal method and space-time processing performance of STAP reverberation can be obtained by low order spatial approximation of the full dimensional data space or to a low dimensional subspace projection, realize dimensionality reduction. According to the covariance matrix of the estimated optimal processing time of large amount of calculation, it is difficult to achieve such problems, the reverberation freedom active homing system are analyzed, and gives the calculation method of reverberation subspace dimension, provides the basis for the study of adaptive dimension reduction. When.
4, on the active homing system dimension STAP reverberation suppression. Research TSA, M-CAP space-time adaptive dimensionality reduction algorithm, the improvement factor, the amount of computation, data sampling, the realization of the complexity analysis of the advantages and disadvantages of drop compared with the optimal processing dimension algorithm. Simulation is given with dimensionality reduction algorithm the space-time processing results and improvement factor curve. Lay the foundation for the study of new STAP suppression method for dimensionality reduction reverberation.
5, propose a method to suppress the reverberation based on information entropy STAP. Based on the study of space-time dimensionality reduction algorithm, introduces the concept of information entropy feature data compression, according to the characteristics of value target and reverberation decomposition showed different space-time data acquisition of data selection, based on the simulation analysis and test.
6, propose a method to suppress the dimension reduction STAP reverberation based on information entropy. The information entropy method in STAP reverberation suppression eigenvalue decomposition caused by increasing the complexity of the algorithm of the algorithm is improved. Based on the information entropy of data extraction, the received data are the number of sub groups, and then short smooth to estimate the covariance matrix, the original space-time processor dimension changed from NK to PQ, which (P N, QK), according to the minimum standards for linearly constrained adaptive weight vector, the target detection data with the size of PQ space-time window smoothing weight vector weighted output.
7, the experimental verification of the reduced dimension STAP reverberation suppression method based on information entropy is carried out. The data from the raw data acquisition to the final output processing are analyzed in detail. The experimental results show that the algorithm can detect the target better while reducing the amount of computation.
8, for any shape of reverberation in the model, the reduced dimension STAP algorithm based on information entropy of arbitrary reverberation suppression. The plane was simulated by the curve equation by the Newton iterative algorithm is calculated in different position of the azimuth and elevation angle. According to the distribution of solution surface equation results obtained by the simulation of the reverberation time. The STAP dimension reduction method based on information entropy calculation of reverberation covariance matrix and weight vector and summarizes the reverberation suppression process. Simulation and analysis verify the seafloor reverberation in arbitrary shape curved surface equation under different level, the output variation and empty Doppler frequency.
9, according to the new surface reverberation model, using STAP dimensionality reduction algorithm based on information entropy of surface reverberation suppression. Based on dynamic surface model, analyzes the influence of different factors on the effect of different factors of surface reverberation and homing system of sea surface reverberation space-time spectrum and power spectrum distribution characteristics. According to the simulation result under the action of different factors of reverberation space-time distribution, the sea is divided into rough surface and rough surface with bubbles two, gives two ways under the sea the space-time processing.

【学位授予单位】：西北工业大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN911.23

【参考文献】