基于规范多元张量分解的矢量传声器阵列定向算法研究  

An Acoustic Vector Sensor Array Orientation Algorithm Based on Tensor Canonical Polyadic Decomposition

作　　者：高宇飞 滕腾 范红波 徐文波 GAO Yufei;TENG Teng;FAN Hongbo;XU Wenbo(The Third Research Institute of China Electronics Technology Group Corporation,Beijing 100015,China)

机构地区：[1]中国电子科技集团公司第三研究所,北京100015

出　　处：《电声技术》2023年第3期42-47,共6页Audio Engineering

基　　金：近地面探测技术重点实验室基金项目(6142414210405)。

摘　　要：针对矢量传声器阵列接收信号模型具有多维结构信息的特性,提出了一种基于规范多元张量分解的阵列定向算法。得益于三阶张量模型结构,相较于传统基于矩阵分解的参数估计方法,该算法可最大化利用接收信号的多维信息,获得更高精度的定向估计性能。通过理论分析证明了该算法可以满足分解唯一性条件,通过数值仿真实验证明了该算法在低信噪比和小快拍数条件下均优于传统定向算法。Aiming at the characteristic that the received signal model of vector microphone aray has multi-dimensional structural information,an array orientation algorithm based on canonical multivariate tensor decomposition is proposed.Thanks to the structure of the third-order tensor model,compared with the traditional parameter estimation method based on matrix decomposition,the algorithm can make maximum use of the multi-dimensional information of the received signal and obtain higher precision direction estimation performance.Theoretical analysis proves that the algorithm can satisfy the decomposition uniqueness condition,and numerical simulation experiments prove that the algorithm is superior to the traditional directional algorithm under the conditions of low signalto-noise ratio and small snapshot number.

关 键 词：矢量传声器阵列 张量分解模型 波达角估计 

分 类 号：TN912.3[电子电信—通信与信息系统] TP212[电子电信—信息与通信工程]