带自适应动量因子的变步长盲源分离方法

doi:10.11959/j.issn.1000-436x.2017067

通信学报 ›› 2017, Vol. 38 ›› Issue (3): 16-24.doi: 10.11959/j.issn.1000-436x.2017067

带自适应动量因子的变步长盲源分离方法

张天骐,马宝泽,强幸子,全盛荣

重庆邮电大学信号与信息处理重庆市重点实验室，重庆 400065

修回日期:2017-01-19 出版日期:2017-03-01 发布日期:2017-04-13
作者简介:张天骐（1971-），男，四川眉山人，博士后，重庆邮电大学教授，主要研究方向为扩频信号的盲处理、神经网络实现以及信号的同步处理。|马宝泽（1990-），男，河北廊坊人，重庆邮电大学硕士生，主要研究方向为盲源分离改进。|强幸子（1986-），男，陕西咸阳人，重庆邮电大学硕士生，主要研究方向为扩频信号盲处理。|全盛荣（1990-），男，湖南衡阳人，重庆邮电大学硕士生，主要研究方向为调频信号的参数估计。
基金资助:
国家自然科学基金资助项目(61671095);国家自然科学基金资助项目(61371164);国家自然科学基金资助项目(61275099);信号与信息处理重庆市级重点实验室建设基金资助项目(CSTC2009CA2003);重庆市教育委员会科研基金资助项目(KJ130524);重庆市教育委员会科研基金资助项目(KJ1600427);重庆市教育委员会科研基金资助项目(KJ1600429)

Variable-step blind source separation method with adaptive momentum factor

Tian-qi ZHANG,Bao-ze MA,Xing-zi QIANG,Sheng-rong QUAN

Chongqing Key Laboratory of Signal and Information Processing,Chongqing University of Posts and Telecommunications,Chongqing 400065,China

Revised:2017-01-19 Online:2017-03-01 Published:2017-04-13
Supported by:
The National Natural Science Foundation of China(61671095);The National Natural Science Foundation of China(61371164);The National Natural Science Foundation of China(61275099);The Project of Key Laboratory of Signal and Information Processing of Chongqing(CSTC2009CA2003);The Research Project of Chongqing Educational Commission(KJ130524);The Research Project of Chongqing Educational Commission(KJ1600427);The Research Project of Chongqing Educational Commission(KJ1600429)

摘要/Abstract

摘要：

基于自然梯度算法提出一种带自适应动量因子的变步长盲源分离方法，在平稳和非平稳环境下进行正定盲源分离处理。该方法利用性能指标构造函数来估计混合矩阵，依据估计混合矩阵得出估计性能指标再反馈更新构造函数；然后将选取合适经验参数的构造函数代入算法，同时自适应调整算法步长和动量因子；最终得到估计源信号。仿真表明该方法在平稳和非平稳环境下都可以估计出混合矩阵，能有效分离混合信号且收敛速度快稳态误差小。

关键词: 盲源分离, 自然梯度, 动量因子, 变步长

Abstract:

A variable-step blind source separation algorithm based on the natural gradient with adaptive momentum factor was proposed,which could cope with the determined blind source separation in the environment of stationary and non-stationary.Function estimation mixed matrix was constructed by performance index.The estimated performance index was obtained by the estimated mixed matrix,and the constructor was updated by the estimated performance index.Then,the constructor was plugged with appropriate experienced parameter into the proposed algorithm and step and momentum factor was adaptively adjusted.Finally,the estimation source signals could be obtained.Simulations show that the proposed algorithm is effective to estimate the mixed matrix in the stationary and non-stationary environments,and the proposed algorithm has faster convergence speed and lower steady error as well as separates source signals effectively.

Key words: blind source separation, natural gradient, momentum factor, variable-step

中图分类号:

TN911.6

张天骐,马宝泽,强幸子,全盛荣. 带自适应动量因子的变步长盲源分离方法[J]. 通信学报, 2017, 38(3): 16-24.

Tian-qi ZHANG,Bao-ze MA,Xing-zi QIANG,Sheng-rong QUAN. Variable-step blind source separation method with adaptive momentum factor[J]. Journal on Communications, 2017, 38(3): 16-24.

图/表 13

图2

图1

图3

图4

图5

图6

图7

图8

图9

图10

表1

图11

表2

参考文献 12

[1]	LIANG Y , HARRIS J , NAQVI S M ,et al. Independent vector analysis with a generalized multivariate Gaussian source prior for frequency domain blind source separation[J]. Signal Processing, 2014,105(12): 175-184.
[2]	季策, 杨坤, 陶奕名 ,等. 一种非平稳环境下的自适应变步长盲源分离算法[J]. 控制与决策, 2016,31(4): 735-739.
	JI C , YANG K , TAO Y M ,et al. An adaptive variable step-size blind source separation algorithm in nonstationary environment[J]. Control and Decision, 2016,31(4): 735-739.
[3]	OU S F , WANG X Y , GAO Y . Adaptive improved RLS algorithm for blind source separation[J]. Wydawnictwo SIGMA-NOT, 2013,89(3b): 81-83.
[4]	欧世峰, 高颖, 赵晓晖 . 自适应组合型盲源分离算法及其优化方案[J]. 电子与信息学报, 2011,33(5): 1243-1247.
	OU S F , GAO Y , ZHAO X H . Adaptive combination algorithm and its modified scheme for blind source separation[J]. Journal of Electronics& Information Technology, 2011,33(5): 1243-1247.
[5]	欧世峰, 高颖, 赵晓晖 . 基于随机梯度的变动量因子自适应白化算法[J]. 自动化学报, 2012,38(8): 1370-1374.
	OU S F , GAO Y , ZHAO X H . Stochastic gradient based variable momentum factor algorithm for adaptive whitening[J]. Acta Automatica Sinica, 2012,38(8): 1370-1374.
[6]	欧世峰, 耿超, 高颖 . 动量项盲源分离算法及其性能优化策略[J]. 电子学报, 2014,42(1): 42-48.
	OU S F , GENG C , GAO Y . Momentum term based blind source separation algorithm and its performance modified strategies[J]. Acta Electronica Sinica, 2014,42(1): 42-48.
[7]	ANDERSON M , ADALI T , LI X L . Joint blind source separation with multivariate gaussian model:algorithms and performance analysis[J]. IEEE Transactions on Signal Processing, 2012,60(4): 1672-1683.
[8]	张贤达 . 矩阵分析与应用[M]. 北京: 清华大学出版社, 2004.
	ZHANG X D . Matrix analysis and applications[M]. Beijing: Tsinghua University PressPress, 2004.
[9]	XU P , SHEN Y , SU Q . Blind source separation with variable step-size method based on a reference separation system[C]// IEEE International Conference on Signal Processing,Communications and Computing. 2014: 110-114.
[10]	YANG H H , AMARI S . Adaptive online learning algorithms for blind separation:maximum entropy and minimum mutual information[J]. Neural Computation, 1997,9(5): 1457-1482.
[11]	MKADEM F , BOUMAIZA S . Physically inspired neural network model for RF power amplifier behavioral modeling and digital predistortion[J]. IEEE Transactions on Microwave Theory ＆ Techniques, 2011,59(4): 913-923．
[12]	陈雷, 张立毅, 郭艳菊 ,等. 基于时间可预测性的差分搜索盲信号分离算法[J]. 通信学报, 2014,35(6): 117-125.
	CHEN L , ZHANG L Y , GUO Y J ,et al. Blind signal separation algorithm based on temporal predictability and differential search algorithm[J]. Journal on Communications, 2014,35(6): 117-125.

算法		Chebysev序列长度			Henon序列长度			Logistic序列长度
算法	500	1 000	1 500	500	1 000	1 500	500	1 000	15 00
FastICA	0.783 5	0.763 5	0.761 3	0.925 8	0.936 8	0.932 1	0.663 6	0.705 5	0.690 7
EASI	0.815 7	0.934 3	0.980 1	0.726 5	0.876 7	0.976 5	0.912 1	0.968 6	0.996 8
VS-M-NGA	0.985 1	0.996 2	0.998 5	0.952 0	0.997 0	0.996 0	0.994 9	0.998 4	0.999 5

算法	Lorentz	Rossler	色噪声
FastICA	0.779 9	0.758 2	0.688 4
EASI	0.943 5	0.951 5	0.919 3
VS-M-NGA	0.966 7	0.950 9	0.996 8

[1]	解元, 邹涛, 孙为军, 谢胜利. 面向高混响环境的欠定卷积盲源分离算法[J]. 通信学报, 2023, 44(2): 82-93.
[2]	马宝泽, 张天骐, 安泽亮, 邓盼. 基于张量分解的卷积盲源分离方法[J]. 通信学报, 2021, 42(8): 52-60.
[3]	陈雷,甘士忠,张立毅,王光艳. 基于样条插值与人工蜂群优化的非线性盲源分离算法[J]. 通信学报, 2017, 38(7): 36-46.
[4]	章鹏,刘全,钟珊,翟建伟,钱炜晟. 增量式双自然策略梯度的行动者评论家算法[J]. 通信学报, 2017, 38(4): 166-177.
[5]	栗华,贾智平,王洪君,刘琚. 基于动态位隙分组盲分离的UHF RFID防碰撞算法[J]. 通信学报, 2012, 33(4): 47-53.
[6]	王荣杰,詹宜巨,周海峰. 基于零空间表示和最大似然的欠定盲源分离方法[J]. 通信学报, 2012, 33(3): 183-190.
[7]	杨小牛,姚俊良,李建东,李钊,张琰. M-PSK/QAM系统中基于相位预编码和盲源分离的共道干扰消除[J]. 通信学报, 2010, 31(8A): 154-160.
[8]	姚俊良,杨小牛,李建东,付卫红,李钊,张琰. 基于最大比合并的超定盲源分离[J]. 通信学报, 2010, 31(7): 9-17.
[9]	宋昕,汪晋宽,刘福来,王彬. 顽健自适应波束形成算法[J]. 通信学报, 2009, 30(6): 7-12.
[10]	张晋东,秦贵和,陈涛,金健. 基于FastICA-TDS的DS-CDMA盲源信号分离系统[J]. 通信学报, 2008, 29(8): 126-130.
[11]	林秋华,党杰,殷福亮. 盲源分离图像加密的相关运算解密法[J]. 通信学报, 2008, 29(1): 109-114.
[12]	郭文强,邱天爽,查代奉. 基于Borel测度峰值判定的欠定混合盲语音信号分离[J]. 通信学报, 2007, 28(9): 22-26.
[13]	谷波,刘琚,许宏吉. 基于独立分量分析的正交空时分组编码盲检测方案[J]. 通信学报, 2006, 27(12): 127-131.

带自适应动量因子的变步长盲源分离方法

Variable-step blind source separation method with adaptive momentum factor

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 12

相关文章 13

Metrics

推荐阅读 0