分数低阶矩的信号盲分离方法

doi:1000-436X(2006)03-0032-05

通信学报 ›› 2006, Vol. 27 ›› Issue (3): 32-36.doi: 1000-436X(2006)03-0032-05

分数低阶矩的信号盲分离方法

张安清^1,²,邱天爽¹,章新华²

¹ 大连理工大学电子与信息工程学院，辽宁大连 116024
² 海军大连舰艇学院信号与信息研究中心，辽宁大连 116018

出版日期:2006-03-25 发布日期:2017-06-22
基金资助:
国家自然科学基金资助项目;国家自然科学基金资助项目

Blind signals separation method based on fractional lower order moments

An-qing ZHANG^1,²,Tian-shuang QIU¹,Xin-hua ZHANG²

¹ School of Electronics and Information Engineering, Dalian University of Technology, Dalian 116024, China
² Research Center of Signal and Information,Dalian Naval Academy, Dalian 116018, China

Online:2006-03-25 Published:2017-06-22
Supported by:
The National Natural Science Foundation of China;The National Natural Science Foundation of China

摘要/Abstract

摘要：

具有尖峰脉冲特性的α稳定分布信号或噪声不存在二阶及二阶以上的统计量，传统基于二阶统计量（SOS）或高阶统计量（HOS）盲分离和独立成份分析（ICA）算法效果不理想，甚至失效。对脉冲噪声环境下的盲分离技术进行了研究，推导出一种基于分数低阶矩的信号盲分离算法，并进行了仿真检验，结果表明所提出的方法对实现α稳定分布信号或含有α稳定分布噪声的信号盲分离效果很好，并具有良好的韧性。

关键词: 分数低阶矩, α稳定分布, 盲分离, 独立成份分析

Abstract:

The impulsive signals or noise with α-stable distribution are characterized by the nonexistence of the finite second or higher order statistics. The traditional blind signals separation (BSS) and ICA approaches were poor perform-ance based on second-order moment or higher-order statistics, even more invalidation. Blind signals separation technique in the presence of impulsive noise was investigated, and a new replace BSS algorithm based on fractional lower order moments was proposed. The algorithm was verified by computer simulation. Simulation results indicate that the new method has good performance to separate signals or noise with α-stable distribution, and very robust.

Key words: fractional lower order moments, α-stable distribution, blind signals separation, independent component analysis

张安清,邱天爽,章新华. 分数低阶矩的信号盲分离方法[J]. 通信学报, 2006, 27(3): 32-36.

An-qing ZHANG,Tian-shuang QIU,Xin-hua ZHANG. Blind signals separation method based on fractional lower order moments[J]. Journal on Communications, 2006, 27(3): 32-36.

图/表 4

参考文献 11

[1]	JUTTEN C , HERAULT J . Blind separation of sources, part I: an adaptive algorithm based on neuromimetic architecture[J]. Signal Processing, 1991,24(1):1-10.
[2]	COMON P . Independent component analysis, a new concept?[J]. Signal Processing, Elsevier, Special Issue on Higher-Order Statistics, 1994,36(1):287-314.
[3]	TEWON L , MARK G , TERRENC E , et al. Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian source[J]. Neural Computation, 1999,11(1):417-441.
[4]	CICHOCKI A , DOUGLAS S C , AMARI S . Robust techniques for independent component analysis (ICA) with noisy data[J]. Neuro-computing, 1998,22:113-129.
[5]	SCHOBBEN D W , SOMMEN P C . A new convolutive blind signal separation algorithm based on second order statistic[A]. Proceedings Int Conf on Signal and Image Proce[C]. 1998.564-569.
[6]	HYVARINEN A . Blind source separation by nonstationarity of vari-ance: a cumulant-based bpproach[J]. IEEE Transactions on Neural Networks, 2001,12(6):1471-1474.
[7]	王惠刚，马远良．水声信号源的盲分离原理及自适应算法[A]． '98年全国声学学术会议论文集．成都， 1998. WANG H G MA Y L . Underwater acoustic signals blind separation principle and adaptive algorithm[A]. '98 Underwater Technical Acoustics Proceedings． Chengdu， 1998.
[8]	倪晋平，马远良，杨坤德．水声信号盲分离的实数与复数算法[J]．西北工业大学学报， 2002,20(1):44-48. NI J P , MA Y L , YANG K D . Underwater acoustic signals blind separation algorithm of real and complex[J]. Journal of Northwestern Ploytechnical University, 2002,20(1):44-48.
[9]	张安清，章新华．基于信息理论的舰船噪声盲分离算法[J]．系统工程与电子技术， 2002,24(1):38-40. ZHANG A Q , ZHANG X H . Based information-theoretic approach to blind separation of ship noise[J]. Systems Engineering and Electronics, 2002,24(1):38-40.
[10]	MOLGEDEY L , SCHUSTER H . Separation of independent signals using time-delayed correlations[J]. Physical Review Letters, 1994,72(23):3634-3637.
[11]	NIKIAS C L , SHAO M . Signal Processing with Alpha-Stable Distributions and Applications[M]. New York: John Wiley and Sons, 1995.

	文献[10]算法		本文算法
	输出1	输出2	输出1	输出2
实验1源信号1	0.7843	0.5126	0.9851	0.0174
实验1源信号2	0.2869	0.6491	0.1786	0.9989
实验2源信号1	0.6711	0.3536	0.9718	0.0580
实验2源信号2	0.4328	0.8189	0.0308	0.9982

[1]	黄强,彭华,李天昀,巩克现. 基于软信息联合修正的非对称PCMA盲分离算法[J]. 通信学报, 2017, 38(4): 178-189.
[2]	郭一鸣,彭华,杨勇,魏雪娟,王浩. 基于DG-PSP的不同符号速率PCMA信号盲分离算法[J]. 通信学报, 2017, 38(3): 92-100.
[3]	贾志成,韩大伟,陈雷,郭艳菊,许浩达. 基于复Givens矩阵与蝙蝠优化的卷积盲分离算法[J]. 通信学报, 2016, 37(7): 107-117.
[4]	杨勇,张冬玲,彭华,涂世龙. 利用Gibbs采样的同频混合信号单通道盲分离[J]. 通信学报, 2015, 36(4): 157-162.
[5]	杜健,巩克现,彭华. 基于CHASE译码思想的PCMA信号盲分离算法[J]. 通信学报, 2015, 36(3): 202-207.
[6]	张冬玲,杨勇,李静,葛临东. 基于Turbo均衡和信道估计的单通道盲信号恢复算法[J]. 通信学报, 2014, 35(1): 47-53.
[7]	刘洋,邱天爽,李景春. 脉冲噪声环境下改进的顽健循环时延估计算法[J]. 通信学报, 2013, 34(6): 184-190.
[8]	廖灿辉，涂世龙，万坚. 抗频偏的突发GMSK混合信号单通道盲分离算法[J]. 通信学报, 2013, 34(5): 10-95.
[9]	廖灿辉,涂世龙,万坚. 抗频偏的突发GMSK混合信号单通道盲分离算法[J]. 通信学报, 2013, 34(5): 88-95.
[10]	邱天爽,李冰,查代奉. 基于分数低阶ICA的混合噪声中脉冲噪声的消除方法[J]. 通信学报, 2011, 32(9): 77-81.
[11]	廖灿辉,涂世龙,万坚. 基于迭代的同频混合信号单通道盲分离/译码算法[J]. 通信学报, 2011, 32(8): 111-117.
[12]	王首勇,朱晓波. 基于FLOC的ARMA SαS模型α谱估计方法[J]. 通信学报, 2007, 28(7): 98-103.
[13]	郭莹,邱天爽,张艳丽,赵勇,栾连毅. 脉冲噪声环境下基于分数低阶循环相关的自适应时延估计方法[J]. 通信学报, 2007, 28(3): 8-14.
[14]	万坚,许华,朱中梁. 基于独立分量分析的成对载波多址分离新算法[J]. 通信学报, 2006, 27(8): 41-44.
[15]	郭莹,邱天爽. 基于分数低阶矩的LETDE时延估计算法[J]. 通信学报, 2006, 27(5): 12-17.

分数低阶矩的信号盲分离方法

Blind signals separation method based on fractional lower order moments

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 11

相关文章 15

Metrics

推荐阅读 0