Algorithm and its performance analysis of principal singular subspace tracking

被引：0

作者：

Du B.-Y. ^{[1
]}

Kong X.-Y. ^{[1
]}

Feng X.-W. ^{[2
]}

机构：

[1] College of Missile Engineering, Rocket Force University of Engineering, Xi'an, 710025, Shaanxi

[2] College of Nuclear Engineering, Rocket Force University of Engineering, Xi'an, 710025, Shaanxi

来源：

Kongzhi Lilun Yu Yingyong/Control Theory and Applications | 2020年 / 37卷 / 07期

基金：

中国国家自然科学基金;

关键词：

Convergence analysis; Neural networks; Principal singular subspace (PSS); Self-stabilizing property;

D O I：

10.7641/CTA.2020.90498

中图分类号：

学科分类号：

摘要：

Principal singular subspace analysis is an adaptive neural network signal processing technique which has been widely applied in modern signal processing. In this paper, a novel information criterion for principal singular subspace tracking is proposed and based on the information criterion an online gradient flow neural network algorithm is derived. Theoretical analysis shows that the information criterion exhibits a unique global minimum point where the state matrices corresponding to the minimum point can exactly span the principal singular subspace of the input signals. The proposed algorithm has a good performance in convergence and an excellent self-stabilizing property. What is more, even if the input signals present a singular cross-correlation characteristic, the proposed algorithm can still track the principal singular subspace of the input signals efficiently. The convergence and self-stability are analyzed via the Lyapunov function approach and ordinary differential equation approach, respectively. MATLAB simulation results verify the effectiveness of the proposed algorithm. © 2020, Editorial Department of Control Theory & Applications South China University of Technology. All right reserved.

引用

页码：1491 / 1500

页数：9

共 25 条

[1]

FENG D Z, ZHANG X D, BAO Z., A neural network learning for adaptively extracting cross-correlation features between two highdimensional data streams, IEEE Transaction on Neural Networks, 15, 6, pp. 1541-1554, (2004)

[2]

OH T, MATSUSHITA Y, TAI Y, Et al., Fast randomized singular value thresholding for low-rank optimization, IEEE Transaction on Pattern Analysis and Machine Intelligence, 40, 2, pp. 376-391, (2017)

[3]

WAN W, ZHOU Z, ZHAO J, Et al., A novel face recognition method: using random weight networks and quasi-singular value decomposition, Neurocomputing, 151, pp. 1180-1186, (2015)

[4]

ZHOU Xinping, SUN Degang, WANG Zhu, Et al., A method based on singular value decomposition for enhancement of differential power analysis, Acta Electronica Sinica, 45, 9, pp. 2250-2255, (2017)

[5]

PANG B, TANG G, TIAN T, Et al., Enhanced singular spectrum decomposition and its application to rolling bearing fault diagnosis, IEEE Access, 7, pp. 87769-87782, (2019)

[6]

XU K, NIE W, FENG D, Et al., A multi-direction virtual array transformation algorithm for 2D DOA estimation, Signal Processing, 125, pp. 122-133, (2016)

[7]

BUNCH J R, NIELSEN C P., Updating the singular value decomposition, Numerical Mathematics, 31, 2, pp. 111-129, (1978)

[8]

MOONEN M, Van DOOREN P, VANDEWALLE J., A singular value decomposition updating algorithm for subspace tracking, Siam Journal on Matrix Analysis and Applications, 13, 4, pp. 1015-1038, (1992)

[9]

CHATTERJEE C, KANG Z, ROYCHOWDHURY V P., Algorithms for accelerated convergence of adaptive PCA, IEEE Transaction on Neural Networks, 11, 2, pp. 338-355, (2000)

[10]

YUILLE A, KAMMEN D, COHEN D., Quadrature and the development of orientation selective cortical cells by Hebb rules, Biological Cybernetics, 61, 3, pp. 183-194, (1989)

← 1 2 3 →