电子学报 ›› 2017, Vol. 45 ›› Issue (1): 29-36.DOI: 10.3969/j.issn.0372-2112.2017.01.005

• 学术论文 • 上一篇    下一篇

一种利用结构特点实现复数域联合对角化解盲源分离新算法研究及应用

徐先峰1, 段晨东1, 刘来君2, 杨小军3   

  1. 1. 长安大学电子与控制工程学院, 陕西西安 710064;
    2. 长安大学公路学院, 陕西西安 710064;
    3. 长安大学信息工程学院, 陕西西安 710064
  • 收稿日期:2015-06-11 修回日期:2015-12-13 出版日期:2017-01-25
    • 作者简介:
    • 徐先峰,男,1982年3月出生于山东宁阳,博士、副教授,硕士生导师.主要研究方向为盲信号处理及其应用.E-mail:xuxianfeng1982@163.com;段晨东,男,1966年4月出生于陕西韩城,博士、教授、硕士生导师.主要研究方向为信号处理,故障诊断与模式识别.E-mail:cdduan@chd.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61201407,No.61473047); 中国博士后科学基金面上资助 (No.2013M542309); 陕西省自然科学基础研究计划 (No.2016JQ5103); 长安大学中央高校基本科研业务费 (No.0009-2014G1321038)

Research on a New Structural Traits Based Complex-Valued Space Joint Diagonalization Algorithm for Blind Source Separation and Its Applications

XU Xian-feng1, DUAN Chen-dong1, LIU Lai-jun2, YANG Xiao-jun3   

  1. 1. School of Electronic & Control Engineering, Chang'an University, Xi'an, Shaanxi 710064, China;
    2. School of Highway, Chang'an University, Xi'an, Shaanxi 710064, China;
    3. School of Information Engineering, Chang'an University, Xi'an, Shaanxi 710064, China
  • Received:2015-06-11 Revised:2015-12-13 Online:2017-01-25 Published:2017-01-25
    • Supported by:
    • National Natural Science Foundation of China (No.61201407, No.61473047); General Program supported by Foundation of China Postdoctoral Science Foundation (No.2013M542309); Natural Science Basic Research Program of Shaanxi Province (No.2016JQ5103); Fundamental Research Funds for the Central Universities of Changan University (No.0009-2014G1321038)

摘要:

联合对角化方法是求解盲源分离问题的有力工具.但是现存的联合对角化算法大都只能求解实数域盲源分离问题,且对目标矩阵有诸多限制.为了求解更具一般性的复数域盲源分离问题,提出了一种基于结构特点的联合对角化(Structural Traits Based Joint Diagonalization,STBJD)算法,既取消了预白化操作解除了对目标矩阵的正定性限制,又允许目标矩阵组为复值,具有极广的适用性.首先,引入矩阵变换,将待联合对角化的复数域目标矩阵组转化为新的具有鲜明结构特点的实对称目标矩阵组.随后,构建联合对角化最小二乘代价函数,引入交替最小二乘迭代算法求解代价函数,并在优化过程中充分挖掘所涉参量的结构特点加以利用.最终,求得混迭矩阵的估计并据此恢复源信号.仿真实验证明与现存的有代表性的对目标矩阵无特殊限制的复数域联合对角化算法FAJD算法及CVFFDIAG算法相比,STBJD算法具有更高的收敛精度,能有效地解决盲源分离问题.

关键词: 盲源分离, 联合对角化, STBJD算法, 交替最小二乘迭代算法

Abstract:

Joint diagonalization (JD) is an efficient tool for blind source separation (BSS) problems.However,most existing JD algorithms could only be used for real-valued space BSS problems and set many constraints on target matrices.In order to solve the general complex-valued space BSS problems,a structural traits based joint diagonalization (STBJD) algorithm is proposed.The algorithm discards pre-whitening procedure,relaxes the positive-definiteness assumption on target matrices and can be used in complex-valued space,thus has more general utilizations.Matrix transformation was adapted to transform the complex-valued space target matrices being jointly diagonalized to real-valued space ones with distinct structural traits.Furthermore,the Least Square cost function for JD was established and solved by alternate least squares (ALS) iterative algorithm.The structural traits of concerned variables were fully exploited and technical utilized in the optimizing process.Finally,the mixing matrix could be estimated and the sources could be retrieved.Numerical simulations illustrated the better convergence performance of STBJD than that of the state-of-the-art algorithms such as FAJD and CVFFDIAG.Thus it could be applied to solve the BSS problems efficiently.

Key words: blind source separation (BSS), joint diagonalization (JD), STBJD algorithm, alternate least squares iterative algorithm

中图分类号: