Characterization of the sources in convolutive mixtures:: A cumulant-based approach

被引：0
作者：
Hornillo-Mellado, S [1 ]
Puntonet, CG
Martín-Clemente, R
Rodríguez-Alvarez, M
Górriz, JM
机构：
[1] Univ Sevilla, Area Teoria Senal & Comunicac, Seville, Spain
[2] Univ Granada, Dept Arquitectura & Tecnol Comp, Granada, Spain
[3] Univ Cadiz, Dept Ingn Sist & Auto Tecn Elect & Elect, Cadiz, Spain
来源：
INDEPENDENT COMPONENT ANALYSIS AND BLIND SIGNAL SEPARATION | 2004年 / 3195卷
关键词：
D O I：
暂无
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
This paper addresses the characterization of independent and non-Gaussian sources in a linear mixture. We present an eigensystem based approach to determine the number of independent components in the signal received by a single sensor. The temporal structure of the sources is also characterized using fourth-order statistics.
引用
收藏
页码：586 / 593
页数：8
相关论文
共 10 条
  • [1] [Anonymous], 2002, Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications
  • [2] [Anonymous], INDEPENDENT COMPONEN
  • [3] ARAKI S, 2003, P IWAENC2003 KYOT JA
  • [4] BENAROYA L, P 4 INT C IND COMP A
  • [5] DOUKAS N, 1996, P 2 WORLD C NONL AN
  • [6] FEVOTTE C, 2003, P 4 INT C IND COMP N
  • [7] Golub G. H., 1996, MATRIX COMPUTATIONS
  • [8] Haykin S., 1991, ADAPTIVE FILTER THEO, DOI DOI 10.1109/ISCAS.2017.8050871
  • [9] LEE TW, 1999, IEEE SIGNAL PROCESSI, V4
  • [10] Nikias C., 1993, Higher-order Spectra Analysis: A Nonlinear Signal Processing Framework
1