| 王翰卓,李风华.随机多项式展开多特征向量约束-匹配场 声源定位算法*[J].,2022,41(4):512-519 |
| 随机多项式展开多特征向量约束-匹配场 声源定位算法* |
| A sound source localization algorithm using multi-eigenvectors constrained matched field processor based on polynomial chaos expansion |
| 投稿时间：2021-07-09 修订日期：2022-06-30 |
| 中文摘要: |
| 多特征向量约束的自适应环境宽容匹配场算法可用于随机和不确知海洋声传播介质中的声源位置定位。该方法需要对拷贝场复声压的互谱矩阵进行估计。为了克服蒙特卡洛统计方法计算互谱矩阵较为耗时的缺点，文章将起伏介质中的随机或不确知声压表示为以正交的随机多项式为基底的级数，随机多项式基底的输入变量为描述环境随机性或不确知性的随机变量。利用随机多项式展开基底的正交性可快速估计拷贝场复声压互谱矩阵。仿真结果表明：在声源频率较低、浅海海水声速存在随机起伏的条件下，在计算效率上使用随机多项式展开方法估计拷贝场复声压互谱矩阵较蒙特卡洛统计方法可提高一个数量级；在高信噪比下，多特征向量约束匹配场声源定位算法在定位准确率和输出峰均比上优于线性匹配场和对角加载的最小方差匹配场声源定位方法。 |
| 英文摘要: |
| The environmentally tolerant multiple-eigenvectors constrained minimal variance matched field processor is a tool for source localization in stochastic and uncertain ocean medium. In order to implement this procedure, the accurate estimation of the cross-spectral density matrix for the complex pressure of the copy filed is needed. And the Monte Carlo statistical method for estimating these matrices is time-consuming. To solve the problem, we approximate the stochastic complex sound pressure as a series of which the bases are orthogonal polynomials. The input variables of these polynomials are random variables which represents the randomness and uncertainties of the environment. The simulation results show that under the condition of low sound frequency and shallow water, as the sound speeds are randomly fluctuated, the method of polynomial chaos expansion improve the computation efficiency by an order of magnitude than the Monte Carlo statistical method for estimating the cross-spectral density matrices. Under the conditions of high input signal-to-noise ratio, the multiple-eigenvectors constrained matched filed source localization algorithm performs better in correct localization rate and outputs’ peak to background ratios than those of the Bartlett and minimal variance matched filed processors. |
| DOI：10.11684/j.issn.1000-310X.2022.04.002 |
| 中文关键词: 匹配场算法，声源定位，随机多项式展开，声传播起伏 |
| 英文关键词: Matched field processor, Source localization, Polynomial chaos expansion, Sound propagation fluctuation |
| 基金项目:国家自然科学基金项目(11674349),国家自然科学基金项目（面上项目，重点项目，重大项目） |
|
| 摘要点击次数: 720 |
| 全文下载次数: 808 |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
