通信学报 ›› 2017, Vol. 38 ›› Issue (1): 106-116.doi: 10.11959/j.issn.1000-436x.2017013

• 学术论文 • 上一篇    下一篇

基于近似核密度估计的近场多声源定位算法

房玉琢,许志勇,赵兆   

  1. 南京理工大学电子工程与光电技术学院,江苏 南京 210094
  • 修回日期:2016-11-30 出版日期:2017-01-01 发布日期:2017-01-23
  • 作者简介:房玉琢(1987-),男,江苏南京人,南京理工大学博士生,主要研究方向为阵列信号处理、声学探测、盲信道辨识等。|许志勇(1968-),男,江苏南京人,博士,南京理工大学副教授,主要研究方向为阵列信号处理、声学探测、雷达技术等。|赵兆(1979-),男,湖北襄阳人,博士,南京理工大学副教授,主要研究方向为声探测系统与信号处理、时频分析。
  • 基金资助:
    国家自然科学基金资助项目(61171167);国家自然科学基金资助项目(61401203);江苏省自然科学基金资助项目(BK20130776)

Near-field localization algorithm of multiple sound sources based on approximated kernel density estimator

Yu-zhuo FANG,Zhi-yong XU,Zhao ZHAO   

  1. School of Electronic and Optical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China
  • Revised:2016-11-30 Online:2017-01-01 Published:2017-01-23
  • Supported by:
    The National Natural Science Foundation of China(61171167);The National Natural Science Foundation of China(61401203);The Natural Science Founda-tion of Jiangsu Province(BK20130776)

摘要:

针对混响环境下的近场多声源定位问题,提出了一种基于近似核密度估计(KDE)的算法模型。引入多阶段(MS)分频带处理有效解决宽间距时的空域模糊,同时,构建空域似然率函数(SLF)通过相加(S)及相乘(P)2种算子进行多维融合,从而衍生出S-KDE、P-KDE、S-KDEMS和P-KDEMS 4种算法。通过对均方根误差(RMSE)以及表征辨识度的SLF百分比(PSLF)这2个统计指标的综合比较,证实了P-KDEMS是一种具有较高稳健性与辨识度的近场多声源定位算法。

关键词: 麦克风阵列, 近似核密度估计, 多阶段分频带处理, 空域似然率函数, 数据融合

Abstract:

For near-field localization of multiple sound sources in reverberant environments,a algorithm model based on approximated kernel density estimator (KDE) was proposed.Multi-stage (MS) of sub-band processing was introduced to effectively solve the spatial aliasing by wide spacing.Spatial likelihood function (SLF) was built for multi-dimensional fusion by using two operators,sum (S) and prod (P).Then four algorithms,S-KDE,P-KDE,S-KDEMS,P-KDEMS,were derived.By the comprehensive comparison of the two statistical indicators root mean square error (RMSE) and percentage of SLF (PSLF) which denoted the recognition,P-KDEMS is confirmed as a near-field localization algorithm of multiple sound sources with high robustness and recognition.

Key words: microphone array, approximated kernel density estimator, multi-stage of sub-band processing, spatial likeli-hood function, data fusion

中图分类号: