利用卷积神经网络的非相干分布式源定位方法  

Incoherently distributed sources localization using convolutional neural network

作　　者：梁奕念 李杰 龙力榕 陈芳炯[1,3] LIANG Yinian;LI Jie;LONG Lirong;CHEN Fangjiong(School of Electronics and Information Engineering,South China University of Technology,Guangzhou 510641;Key Laboratory of Ocean Observation Technology,Ministry of Natural Resources,Tianjin 300112;Guangdong Provincial Key Laboratory of Short-Range Wireless Detection and Communication,Guangzhou 510640)

机构地区：[1]华南理工大学电子与信息学院,广州510641 [2]自然资源部海洋观测技术重点实验室,天津300112 [3]广东省短距离无线探测与通信重点实验室,广州510640

出　　处：《声学学报》2024年第1期38-48,共11页Acta Acustica

基　　金：国家自然科学基金项目(61971198,62271208);自然资源部海洋观测技术重点实验室开放基金项目(2021klootA05);广东省基础与应用基础研究基金项目(2022A1515011305);广东省科技计划项目(2022A0505050011)资助。

摘　　要：针对传统子空间方法对分布式源定位依赖模型假设以及子空间有效维度难以选择的问题,提出了一种利用卷积神经网络的非相干分布式源定位方法。该方法把卷积神经网络作为一个强鲁棒性空间功率密度分布特征提取器,实现从协方差矩阵到方向角功率密度分布的映射。根据得到的空间谱分布,可进一步实现分布式源的参数估计。此外,文中结合迁移学习技术解决实际信号源分布与训练模型不匹配的问题,提升了模型的泛化性能。仿真实验表明该方法对不同信号分布模型具有稳健性,参数估计性能优于传统子空间方法。传声器阵列实测数据表明该方法的中心角和角度扩展的估计误差在1°以内。To solve the problem that the traditional subspace methods for incoherent distributed sources location are difficult to select the effective dimension of the subspace,and rely on the model assumption,an incoherently distributed source localization method based on convolutional neural networks is proposed.As a robust spatial power density distribution feature extractor,convolutional neural networks realize the mapping from the covariance matrix to the direction angle power density distribution.On this basis,the key parameters can also be extracted from the estimated spatial spectrum.In addition,transfer learning technique is employed to solve the mismatch problem between the real signal source distribution and the training model,and improve the generalization performance of the model.Simulation results demonstrate the proposed method is robust to different distributed source models and has better parameter estimation performance than the traditional subspace methods.The real data from microphone array shows that the estimation error of the central angle and the distributed angle with this method is less than 1°.

关 键 词：波达方向估计 角度扩展估计 非相干分布式源 深度学习 阵列信号处理 

分 类 号：TN911.7[电子电信—通信与信息系统] TP183[电子电信—信息与通信工程]