Kernel Recursive Least Squares With Multiple Feedback and Its Convergence Analysis

被引：16

作者：

Wang, Shiyuan ^{[1
,2
]}

Wang, Wanli ^{[1
,2
]}

Duan, Shukai ^{[1
,2
]}

Wang, Lidan ^{[1
,2
]}

机构：

[1] Southwest Univ, Coll Elect & Informat Engn, Chongqing 400715, Peoples R China

[2] Chongqing Key Lab Nonlinear Circuits & Intelligen, Chongqing 400715, Peoples R China

来源：

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS | 2017年 / 64卷 / 10期

基金：

中国国家自然科学基金;

关键词：

Convergence; feedback network; kernel recursive least squares (KRLS); steady state mean square error (MSE); MEAN-SQUARE; ALGORITHM; PROJECTION; SYSTEMS;

D O I：

10.1109/TCSII.2017.2654263

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

In kernel adaptive filters (KAFs), feedback network can lead to performance improvement from the aspects of estimation accuracy and convergence rate. In this brief, a novel feedback structure is developed and applied to the kernel recursive least squares (KRLS), generating the KRLS with multiple feedback (KRLS-MF). In the proposed KRLS-MF, multiple previous outputs are utilized to update the structure parameters in the recurrent form. The obtained parameters are also proved to be convergent. Compared with other KAFs with and without feedback, KRLS-MF can improve both the filtering accuracy and convergence rate, efficiently.

引用

页码：1237 / 1241

页数：5

共 25 条

[1]

[Anonymous], 2009, Kernel methods for remote sensing data analysis

[2] Generalized Correntropy for Robust Adaptive Filtering [J].

Chen, Badong ;

Xing, Lei ;

Zhao, Haiquan ;

Zheng, Nanning ;

Principe, Jose C. .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2016, 64 (13) :3376-3387

[3] Quantized Kernel Recursive Least Squares Algorithm [J].

Chen, Badong ;

Zhao, Songlin ;

Zhu, Pingping ;

Principe, Jose C. .

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2013, 24 (09) :1484-1491

[4] ROBUSTNESS ANALYSIS OF DISCRETE-TIME ADAPTIVE-CONTROL SYSTEMS USING INPUT-OUTPUT STABILITY THEORY - A TUTORIAL [J].

CLUETT, WR ;

SHAH, SL ;

FISHER, DG .

IEE PROCEEDINGS-D CONTROL THEORY AND APPLICATIONS, 1988, 135 (02) :133-141

[5] The kernel recursive least-squares algorithm [J].

Engel, Y ;

Mannor, S ;

Meir, R .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2004, 52 (08) :2275-2285

[6] Online learning with kernel regularized least mean square algorithms [J].

Fan, Haijin ;

Song, Qing ;

Shrestha, Sumit Bam .

KNOWLEDGE-BASED SYSTEMS, 2014, 59 :21-32

[7] An information theoretic sparse kernel algorithm for online learning [J].

Fan, Haijin ;

Song, Qing ;

Xu, Zhao .

EXPERT SYSTEMS WITH APPLICATIONS, 2014, 41 (09) :4349-4359

[8] A linear recurrent kernel online learning algorithm with sparse updates [J].

Fan, Haijin ;

Song, Qing .

NEURAL NETWORKS, 2014, 50 :142-153

[9] A neural-network-based channel-equalization strategy for chaos-based communication systems [J].

Feng, JC ;

Tse, CK ;

Lau, FCM .

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2003, 50 (07) :954-957

[10] An Affine Projection Algorithm With Periodically Evolved Update Interval [J].

Kim, Kwang-Hoon ;

Choi, Young-Seok ;

Kim, Seong-Eun ;

Song, Woo-Jin .

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2011, 58 (11) :763-767

← 1 2 3 →