张伟涛,,
楼顺天
西安电子科技大学电子工程学院 西安 710071
基金项目:国家自然科学基金(61571339)，陕西省创新人才推进计划-青年科技新星项目(2018KJXX-019)
详细信息
作者简介:李扬：男，1987年生，博士生，研究方向为盲信号处理
张伟涛：男，1983年生，副教授，硕士生导师，研究方向为盲信号处理、语音信号处理
楼顺天：男，1962年生，教授，博士生导师，研究方向为神经网络信息处理与应用、模糊信息处理与应用、盲信号处理、现代信号智能处理、智能控制技术
通讯作者:张伟涛 zhwt-work@foxmail.com
中图分类号:TN911.7计量
文章访问数:2230
HTML全文浏览量:1208
PDF下载量:36
被引次数:0
出版历程
收稿日期:2019-01-24
修回日期:2019-06-11
网络出版日期:2019-06-24
刊出日期:2019-12-01
Deep Convolution Blind Separation of Acoustic Signals Based on Joint Diagonalization
Yang LI,Weitao ZHANG,,
Shuntian LOU
Institute of Electronic Engineering, Xidian University, Xi’an 710071, China
Funds:The National Natural Science Foundation of China (61571339), The Innovative Talents Promotion Program of Shaanxi Province (2018KJXX-019)
摘要
摘要:声信号在空间中的传播具有较强的多径效应，在接收端往往以卷积形式相互叠加，尤其在海洋、剧场等强混响条件下，混合滤波器冲激响应的长度会显著增加，现有的频域卷积盲分离算法将失效。为了消除长脉冲响应导致解混合模型失效的问题，该文对观测信号进行两次短时傅里叶变换(STFT)，第1次STFT缩短了脉冲响应长度，第2次STFT将信号模型转化为瞬时盲分离，最终利用联合对角化(JD)技术估计出分离矩阵。与现有方法相比，所提方法解决了深度卷积混合下模型失效的问题，并且当源信号数较多或存在加性噪声时，可以得到更好的分离性能。仿真结果验证了方法的有效性和性能优势。
关键词:盲源分离/
深度卷积/
联合对角化/
排序问题
Abstract:The propagation of acoustic signal in space has a strong multipath effect, and the receiver often overlaps in the form of convolution. Especially in strong reverberation conditions such as ocean and theatre, where the length of impulse response of hybrid filter increases significantly. In order to eliminate the problem that long impulse response leads to the failure of the frequency domain convolution blind separation algorithm, two Short-Time Fourier Transforms (STFT) are applied to the observed signal. The first STFT shortens the length of the hybrid filter. The second STFT converts the signal model into instantaneous blind separation. Finally, the separation matrix is estimated by Joint Diagonalization (JD) technique. Compared with the existing methods, this method solves the problem of model failure under deep convolution mixing, and can obtain better separation performance when the number of source signals is large or additive noise exists. The simulation results verify the effectiveness and performance advantages of the proposed method.
Key words:Blind source separation/
Deep convolution/
Joint Diagonalization (JD)/
Permutation problem
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=a43d4e96-d2d3-424d-b2c5-e6846b746856
