匹配场反演阵形估计算法的试验研究

传统的阵形估计算法都基于这样一个假设,即声传播为平面波或球面波模型,在浅海应用条件下,多途效应显著,该假设失效,造成此类算法的性能下降。提出了一种基于匹配场反演的阵形估计算法,采用简正波模型对声场建模,将多途建模为不同号数的简正波,从而消除了多途效应的影响;并且通过合理地选择目标函数,使算法对环境参数的扰动具有一定的宽容性;利用柔性阵的线性约束,将阵形畸变用弧线建模,使问题简化,在缺少声源精确位置等先验知识的前提下,仍能实现阵形的准确估计。理论仿真和试验结果表明,该算法是有效的。

Array shape estimation based on matched field inversion

Planar or sphere model is often used in traditional array shape estimation methods.Due to suffering from multipath effect in shallow water applications,the performance of these methods is degraded seriously.An algorithm based on Matched Field Inversion（MFI） is introduced in this paper.This algorithm applies Normal Mode assumption so that the multipath effect can be eliminated.Given a reasonable choice of the target function this algorithm has excellent tolerance of environmental perturbations.And the complexity is reduced by modeling the perturbed array shape with a piece of arc.Theoretical simulation and experimental results show the efficient of the algorithm.