次成分提取信息准则的加权规则方向收敛分析

doi:10.11959/j.issn.1000-436x.2020014

通信学报 ›› 2020, Vol. 41 ›› Issue (3): 25-32.doi: 10.11959/j.issn.1000-436x.2020014

次成分提取信息准则的加权规则方向收敛分析

杜柏阳¹,孔祥玉¹,冯晓伟²

¹ 火箭军工程大学导弹工程学院，陕西西安 710025
² 火箭军工程大学核工程学院，陕西西安 710025

修回日期:2019-12-15 出版日期:2020-03-25 发布日期:2020-03-31
作者简介:杜柏阳（1990– ），男，山东滨州人，火箭军工程大学博士生，主要研究方向为信号特征提取、大型工业过程故障诊断|孔祥玉（1967– ），男，山西临汾人，博士，火箭军工程大学教授、博士生导师，主要研究方向为信号特征提取、复杂非线性系统建模、大型工业过程故障监测与故障诊断|冯晓伟（1986– ），男，四川绵阳人，博士，火箭军工程大学讲师，主要研究方向为信号特征提取、大型工业过程故障监测与故障诊断
基金资助:
国家自然科学基金资助项目(61673387);国家自然科学基金资助项目(61374120);国家自然科学基金资助项目(61903375);国家自然科学基金资助项目(61833016)

Direction convergence analysis of weighted rule for minor component extraction information criteria

Boyang DU¹,Xiangyu KONG¹,Xiaowei FENG²

¹ Department of Missile Engineering,the Rocket Force University of Engineering,Xi’an 710025,China
² Department of Nuclear Engineering,the Rocket Force University of Engineering,Xi’an 710025,China

Revised:2019-12-15 Online:2020-03-25 Published:2020-03-31
Supported by:
The National Natural Science Foundation of China(61673387);The National Natural Science Foundation of China(61374120);The National Natural Science Foundation of China(61903375);The National Natural Science Foundation of China(61833016)

摘要/Abstract

摘要：

对并行次成分提取算法在信号特征提取中实现方向收敛的过程进行了研究。采用比较未加权的PAST次子空间跟踪算法和加权的PAST并行次成分提取算法，分析加权规则对次成分提取算法方向收敛行为的限定方式。理论分析表明，加权规则对状态矩阵的各个向量和次成分之间的角度限定存在规律，并给出了加权规则对状态矩阵各个向量方向的作用方式和定理和讨论。最后，Matlab仿真实验验证了理论分析的结果。

关键词: 次子空间提取算法, 加权规则, 并行次成分分析, 信息准则

Abstract:

The extraction of parallel minor components algorithm in directional convergence in signal feature extraction was studied.By comparing the unweighted projection approximation subspace tracking (PAST) algorithm with the weighted PAST parallel minor component extraction algorithm,the evolution method of the minor component extraction algorithm was analyzed.Theorical analysis illustrated that the weighted rule was able to guide the angle evolution between the vectors of the state matrix and minor components.Finally,Matlab simulation verifies the validity of the proposed theory.

Key words: minor subspace extraction algorithm, weighted rule, parallel minor component analysis

中图分类号:

TP273

杜柏阳,孔祥玉,冯晓伟. 次成分提取信息准则的加权规则方向收敛分析[J]. 通信学报, 2020, 41(3): 25-32.

Boyang DU,Xiangyu KONG,Xiaowei FENG. Direction convergence analysis of weighted rule for minor component extraction information criteria[J]. Journal on Communications, 2020, 41(3): 25-32.

图/表 8

图1

图2

图3

图4

图5

图6

图7

图8

参考文献 16

[1]	GAO Y B , KONG X Y , HU C H ,et al. Convergence analysis of M?ller algorithm for estimating minor component[J]. Neural Processing Letters, 2015,42(2): 268-355.
[2]	KONG X Y , HU C H , HAN C Z . A self-stabilizing MSA algorithm in high-dimensional data stream[J]. Neural Networks, 2010,23(7): 865-871.
[3]	KONG X Y , HU C H , HAN C Z . A dual purpose principal and minor subspace gradient flow[J]. IEEE Transactions on Signal Processing, 2012,60(1): 197-210.
[4]	NGUYEN D T , YAMADA I . A unified convergence analysis of normalized PAST algorithms for estimating principal and minor components[J]. Signal Processing, 2013,93(1): 176-184.
[5]	M?LLER R . A self-stabilizing learning rule for minor component analysis[J]. International Journal of Neural Systems, 2004,14(1): 1-8.
[6]	CIRRINCIONE G , CIRRINCIONE M . Neural-based orthogonal data fitting:the EXIN neural networks[J]. Aevum, 2010(2): 368-371.
[7]	GAO Y B , KONG X Y , ZHANG H H ,et al. A weighted information criterion for multiple minor components and its adaptive extraction algorithms[J]. Neural networks, 2017,89(5): 1-10.
[8]	YANG B . Projection approximation subspace tracking[J]. IEEE transactions on signal processing, 1995,43(1): 95-107.
[9]	FENG D Z , ZHENG W , JIA Y . Neural network learning algorithms for tracking minor subspace in high-dimensional data stream[J]. IEEE Transactions on Neural Networks, 2005,16(3): 513-521.
[10]	JANKOVIC M , OGAWA H . Time-oriented hierarchical method for computation of minor components[M]. Berlin: SpringerPress, 2005.
[11]	TAN K K , LYU J , ZHANG Y ,et al. Adaptive multiple minor directions extraction in parallel using a PCA neural network[J]. Theoretical Computer Science, 2010,411(48): 4200-4215.
[12]	THAMERI M , ABED-MERAIM K , BELOUCHRANI A . Low complexity adaptive algorithms for principal and minor component analysis[J]. Digital Signal Processing, 2013,23(1): 19-29.
[13]	OUYANG S , BAO Z . Fast principal component extraction by a weighted information criterion[J]. IEEE Transactions on Signal Processing, 2002,50(8): 1994-2002.
[14]	BARTELMAOS S , ABED-MERAIM K . Fast adaptive algorithms for minor component analysis using Householder transformation[J]. Digital Signal Processing, 2011,21(6): 667-678.
[15]	TOSHIHISA T . Generalized weighted rules for principal components tracking[J]. IEEE Transactions on Signal Processing, 2005,53(4): 1243-1253.
[16]	ZHANG Y , YE M , LYU J C ,et al. Convergence analysis of a deterministic discrete time system of Oja’s PCA learning algorithm[J]. IEEE Transactions on Neural Networks, 2005,16(6): 1318-1328.

次成分提取信息准则的加权规则方向收敛分析

Direction convergence analysis of weighted rule for minor component extraction information criteria

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 16

相关文章 1

Metrics

推荐阅读 0