| 引用本文： |
-
邢传玺,谈光枝,冉艳玲,等.一种嵌合互质阵列的离网水声信号方位估计方法[J].电讯技术，2026，（4）：681 - 689. [点击复制]
- XING Chuanxi,TAN Guangzhi,RAN Yanling,et al.An Off-grid Hydroacoustic Signal Orientation Estimation Method for Chimeric Coprime Arrays[J].，2026，（4）：681 - 689. [点击复制]
|
|
| 摘要: |
| 针对水听器互质阵列存在无法有效利用全部阵元信息和海洋噪声影响下传统算法波达方向（Direction of Arrival，DOA）估计性能下降的问题，提出了一种结合互质阵列内插值及信号子空间拟合的离格稀疏贝叶斯水声信号方位估计方法。针对水下声信号接收与方位估计问题，采用互质阵列结构获取水声观测数据，通过虚拟阵元插值策略对虚拟域中存在的阵元间隙进行有效填补。基于原子范数最小化优化方法，完成插值后虚拟阵列协方差矩阵的精确重构，实现对阵列接收信号全域信息的充分利用，为DOA估计奠定数据基础。利用重构协方差矩阵所对应的信号子空间对原始接收数据实施信号重构，可显著削弱水下复杂背景噪声对估计结果的不利影响。结合水声信号场固有的空域稀疏分布特征，建立重构信号的离格稀疏表征模型，引入贝叶斯学习机制求解信源信号的最大后验概率估计值，进而完成水下目标方位的精准解算。仿真实验结果表明，所提方法仅采用7个物理阵元即可实现对12个信源方位角的有效估计；当信噪比低至-20 dB时，该方法相较于互质内插值阵列架构下的ROOTMUSIC算法，估计精度提升幅度达到46.89%；在快拍数设置为256的条件下，其估计性能较对比算法提升22.12%。 |
| 关键词: 水声信号 DOA估计 内插阵元 互质阵列 子空间信号重构 离格稀疏贝叶斯推理 |
| DOI：10.20079/j.issn.1001-893x.240424004 |
|
| 基金项目:国家自然科学基金资助项目（61761048）；云南省基础研究专项面上项目（20210AT070132） |
|
| An Off-grid Hydroacoustic Signal Orientation Estimation Method for Chimeric Coprime Arrays |
| XING Chuanxi,TAN Guangzhi,RAN Yanling,LU Mao,MENG Qiang |
| (1.College of Electrical and Information Technology，Yunnan Minzu University，Kunming 650500，China;2.Yunnan Key Laboratory of Unmanned Autonomous System，Kunming 650504，China) |
| Abstract: |
| For the problem that hydrophone coprime arrays cannot effectively exploit all array element information and the traditional algorithm’s direction of arrival(DOA) estimation performance degrades under the influence of ocean noise，a method combining coprime array interpolation and signal subspace approximation combined with off-grid sparse Bayesian inference method is proposed.First，a coprime array is used to receive underwater acoustic signals，and the missing holes in the virtual domain are filled by interpolating virtual array elements.By minimizing the design of atomic norms and reconstructing the covariance matrix of the interpolated virtual array，all signals received by the array are information used for DOA estimation. Second，the received signal is reconstructed by the reconstructed covariance signal subspace，which effectively reduces the background noise interference.Finally，spatial sparsity is used to derive an off-lattice sparse representation model of the reconstructed signal，and Bayesian learning is used to compute the maximum posterior probability of the source signal to achieve target orientation estimation.Simulation shows that 12 source angles can be effectively estimated using sub-arrays with array elements of 7，and that DOA estimation can be performed accurately under low signal-to-noise ratio(SNR).When the SNR is -20 dB，the proposed algorithm improves the estimation performance by 46.89% compared with the ROOT Modified Multiple Signal Classification(ROOTMUSIC) algorithm under the coprime interpolation array.When the number of snapshots is 256，the proposed algorithm improves the estimation performance by 2212% compared with the ROOTMUSIC algorithm under the coprime interpolation array. |
| Key words: hydroacoustic signal DOA estimation interpolation elements coprime array subspace signal reconstruction off-grid sparse Bayesian inference |
《电讯技术》编辑部 