基于非线性主成分分析的自适应变步长盲源分离算法

doi:10.3724/SP.J.1087.2013.01233

计算机应用 ›› 2013, Vol. 33 ›› Issue (05): 1233-1236.DOI: 10.3724/SP.J.1087.2013.01233

基于非线性主成分分析的自适应变步长盲源分离算法

辜方林¹,张杭¹,李伦辉²

1. 解放军理工大学通信工程学院,南京 210007
2. 中国人民解放军75708部队,长沙 410007

收稿日期:2012-09-17 修回日期:2012-11-06 出版日期:2013-05-01 发布日期:2013-05-08
通讯作者: 辜方林
作者简介:辜方林(1986-),男,湖南新化人,博士研究生,主要研究方向:盲信号处理、统计信号处理;张杭(1962-),女,浙江天台人,教授,博士生导师,主要研究方向:盲信号处理、认知无线电、卫星通信;李伦辉(1983-),男,湖南衡阳人,硕士,主要研究方向:信号与信息处理。
基金资助:
国家自然科学基金资助项目(61261039);国家973计划项目(2009CB320400)

Adaptive variable step-size blind source separation algorithm based on nonlinear principal component analysis

GU Fanglin¹,ZHANG Hang¹,LI Lunhui²

1. College of Communication Engineering, PLA University of Science and Technology, Nanjing Jiangsu 210007, China
2. No. 75708 Troops of PLA, Changsha Hunan 410007, China

Received:2012-09-17 Revised:2012-11-06 Online:2013-05-08 Published:2013-05-01
Contact: GU Fanglin

摘要/Abstract

摘要： 算法的迭代步长对于算法的收敛性能有着重要影响。针对固定步长的非线性主成分分析(NPCA)算法不能兼顾收敛速度和估计精度的情形，提出基于梯度的自适应变步长NPCA算法和最优变步长NPCA算法两种自适应变步长算法来改善其收敛性能。特别地,最优变步长NPCA算法通过对代价函数进行一阶线性近似表示,从而计算出当前的最优迭代步长。该算法的迭代步长随估计误差的变化而变化,估计误差大,迭代步长相应大,反之亦然;且不需要人工设置任何参数。仿真结果表明,当算法的估计精度相同时,与固定步长NPCA算法相比,两种自适应变步长NPCA算法相对固定步长NPCA算法都具有更好的收敛速度或跟踪性能,且最优变步长NPCA算法的性能优于基于梯度的自适应变步长NPCA算法。

关键词: 盲源分离, 非线性主成分分析, 变步长

Abstract: The design of the step-size is crucial to the convergence rate of the Nonlinear Principle Component Analysis (NPCA) algorithm. However, the commonly used fixed step-size algorithm can hardly satisfy the convergence speed and estimation precision requirements simultaneously. To address this issue, the gradient-based adaptive step-size NPCA algorithm and optimal step-size NPCA algorithm were proposed to speed up the convergence rate and improve tracking ability. In particular, the optimal step-size NPCA algorithm linearly approximated the contrast function and figured out the optimal step-size currently. The optimal step-size NPCA algorithm utilized an adaptive step-size whose value was adjusted in sympathy with the value of the contrast function and free from any manual parameters. The simulation results show that the proposed adaptive step-size NPCA algorithms have faster convergence rate or better tracking ability in comparison with the fixed step-size NPCA algorithm when the estimation precisions are same. The convergence performance of the optimal step-size NPCA algorithm is superior to that of the gradient-based adaptive NPCA algorithm.

Key words: blind source separation, Nonlinear Principal Component Analysis (NPCA), variable step-size

中图分类号:

TN911.7

辜方林张杭李伦辉. 基于非线性主成分分析的自适应变步长盲源分离算法[J]. 计算机应用, 2013, 33(05): 1233-1236.

GU Fanglin ZHANG Hang LI Lunhui. Adaptive variable step-size blind source separation algorithm based on nonlinear principal component analysis[J]. Journal of Computer Applications, 2013, 33(05): 1233-1236.

参考文献

[1]罗双才,唐斌.一种基于盲分离的欺骗干扰抑制算法[J]. 电子与信息学报,2011,33(12):2801-2806.

[2]WANG L, DING H, YIN F. Target speech extraction in cocktail party by combining beamforming and blind source separation[J]. Acoustics Australia, 2011, 39(2): 64-68.

[3]王宇,迟毅林,伍星,等.基于盲信号处理的机械噪声监测和故障诊断[J].振动与冲击,2009,28(6):32-41.

[4]COMON P. Independent component analysis, a new concept?[J]. Signal Processing, 1994, 36(3): 287-314.

[5]BELL A J, SEJNOWSKI T J. An information-maximization approach to blind separation and blind deconvolution[J]. Neural Computation, 1995, 7(6): 1129-1159.

[6]HYVARINEN A, OJA E. A fast fixed-point algorithm for independent component analysis[J]. Neural Computation, 1997, 9(7): 1483-1492.

[7]CAIAFA C F, PROTO A N. Separation of statistically dependent sources using an l2-distance non-Gaussianity measure[J]. Signal Processing, 2006, 86(11): 3404-3420.

[8]OJA E. The nonlinear PCA learning rule in independent component analysis[J]. Neurocomputing, 1997, 17(1): 25-45.

[9]蒋照菁，辜方林，张杭. 一种基于NPCA的自适应变步长盲源分离算法[J]. 计算机工程与应用，2013，49(8)：206-208.

[10]NAKAJIMA H, NAKADAI K, HASEGAWA Y, et al. Adaptive step-size parameter control for real-world blind source separation[C]// ICASSP 2008: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Piscataway, NJ: IEEE Press, 2008: 149-152.

[11]NAKAJIMA H, NAKADAI K, HASEGAWA Y, et al. Blind source separation with parameter-free adaptive step-size method for robot audition[J]. IEEE Transactions on Audio, Speech, and Language Processing, 2010, 18(6): 1476-1485.

[12]NORDHAUSEN K, OLLILA E, OJA H. On the performance indices of ICA and blind source separation[C]// Proceedings of IEEE 12th International Workshop on Signal Processing Advances in Wireless Communications. Piscataway, NJ: IEEE Press, 2011: 461-465.

[1]	王晓静, 彭虎, 邓长寿, 黄海燕, 张艳, 谭旭杰. 基于均匀局部搜索和可变步长的萤火虫算法[J]. 计算机应用, 2018, 38(3): 715-721.
[2]	郑奕丰, 郝学元, 颜晓红. 基于水声通信的新型变步长最小均方算法[J]. 计算机应用, 2017, 37(8): 2195-2199.
[3]	叶卫东, 杨涛. 融合极点对称模态分解与时频分析的单通道振动信号盲分离方法[J]. 计算机应用, 2016, 36(10): 2933-2939.
[4]	张光宇, 陈红, 蔡晓霞. 基于RobustICA的数字调制混合信号盲源分离算法[J]. 计算机应用, 2015, 35(8): 2129-2132.
[5]	黄艳艳, 彭华. 分布式MIMO-OFDM信号多频偏多信道联合盲估计[J]. 计算机应用, 2015, 35(6): 1531-1536.
[6]	穆宇超, 张小红. 自适应树形分组的盲分离射频识别系统防碰撞算法[J]. 计算机应用, 2015, 35(1): 19-22.
[7]	李丽娜曾庆勋甘晓晔梁德骕. 基于势函数与压缩感知的欠定盲源分离[J]. 计算机应用, 2014, 34(3): 658-662.
[8]	韩亮蒲秀娟. 使用时频盲源分离和小波包去噪的胎儿心电信号提取[J]. 计算机应用, 2013, 33(08): 2394-2396.
[9]	王瑜琳田学隆高雪利. 基于解相关变步长的改进型语音增强算法[J]. 计算机应用, 2013, 33(06): 1746-1749.
[10]	张艳萍纪磊. 基于指数型多延迟误差信号自相关的变步长常模算法[J]. 计算机应用, 2013, 33(03): 625-627.
[11]	程园园李海燕陈海涛施心陵. 灰度极限脉冲耦合神经网络混合噪声滤波[J]. 计算机应用, 2012, 32(03): 729-731.
[12]	王华秋刘轲. 改进CMAC在森林火焰识别中的应用[J]. 计算机应用, 2011, 31(03): 860-864.
[13]	陈永萍朱庆生葛垚. 结合变步长的两阶段视频镜头边界检测方法[J]. 计算机应用, 2008, 28(6): 1501-1503.
[14]	朱坚坚王惠刚李虎雄. 联合频域盲语音分离排序算法[J]. 计算机应用, 2008, 28(6): 1552-1554.
[15]	王英志. 一种波束域子带盲源分离新方法[J]. 计算机应用, 2008, 28(11): 2987-2990.

基于非线性主成分分析的自适应变步长盲源分离算法

Adaptive variable step-size blind source separation algorithm based on nonlinear principal component analysis

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics