Theoretical Analysis of LMS Algorithm for Time-Varying Unknown System

被引：0

作者：

Ishibushi, Norihiro ^{[1
]}

Kajikawa, Yoshinobu ^{[2
]}

Miyoshi, Seiji ^{[2
]}

机构：

[1] Kansai Univ, Grad Sch Sci & Engn, Suita, Osaka, Japan

[2] Kansai Univ, Fac Engn Sci, Suita, Osaka, Japan

来源：

2016 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA) | 2016年

关键词：

ADAPTIVE FILTER; LEARNING CHARACTERISTICS;

D O I：

暂无

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：

We analyze the behaviors of the LMS algorithm for a time-varying unknown system using a statistical-mechanical method. We obtain simultaneous differential equations that describe the dynamical behaviors of the macroscopic variables under conditions in which the tapped-delay line is sufficiently long. We show the existence of an optimal step size owing to the trade-off between the noise misadjustment and the lag misadjustment. Furthermore, we obtain the exact optimal step size in the case of a white reference signal. The derived theory includes the behaviors for a time-constant unknown system as a special case.

引用

页数：4

共 50 条

[1] Statistical-Mechanical Analysis of LMS Algorithm for Time-Varying Unknown System
Ishibushi, Norihiro
Kajikawa, Yoshinobu
Miyoshi, Seiji
[J]. JOURNAL OF THE PHYSICAL SOCIETY OF JAPAN, 2017, 86 (02)
[2] Time-varying system identification using the quantized LMS algorithm
Bermudez, JCM
Bershad, NJ
[J]. 38TH MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS, PROCEEDINGS, VOLS 1 AND 2, 1996, : 247 - 250
[3] Transient and tracking performance analysis of the quantized LMS algorithm for time-varying system identification
Bermudez, JCM
Bershad, NJ
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1996, 44 (08) : 1990 - 1997
[4] Performance of the LMS algorithm on the estimation of time-varying channels
Galdino, JF
Pinto, EL
de Alencar, MS
[J]. 2002 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS, VOLS 1-5, CONFERENCE PROCEEDINGS, 2002, : 94 - 98
[5] CONTROL ALGORITHM FOR UNKNOWN TIME-VARYING SYSTEMS
PENFOLD, HB
EVANS, RJ
[J]. INTERNATIONAL JOURNAL OF CONTROL, 1989, 50 (01) : 13 - 32
[6] Performance analysis of a New Time-Varying LMS(NTVLMS) Adaptive Filtering Algorithm
Siddappaji
Sudha, K. L.
[J]. FIRST INTERNATIONAL CONFERENCE ON EMERGING TRENDS IN ENGINEERING, TECHNOLOGY AND SCIENCE - ICETETS 2016, 2016,
[7] Analysis of Linearity and Time-Varying Characteristic of Adaptive Detecting Distorted Current System Based on LMS Algorithm
Liu, Chuanlin
Liu, Kaipei
[J]. 2009 ASIA-PACIFIC POWER AND ENERGY ENGINEERING CONFERENCE (APPEEC), VOLS 1-7, 2009, : 3191 - 3195
[8] Analyze Linearity and Time-Varying Characteristic of Continuous Detecting Distorted Current System Based on LMS Algorithm
Liu, Kaipei
Liu, Chuanlin
[J]. 2009 ASIA-PACIFIC POWER AND ENERGY ENGINEERING CONFERENCE (APPEEC), VOLS 1-7, 2009, : 3196 - 3200
[9] NONLINEAR ADAPTIVE LEARNING CONTROL FOR UNKNOWN TIME-VARYING PARAMETERS AND UNKNOWN TIME-VARYING DELAYS
Chen, Weisheng
Zhang, Zhengqiang
[J]. ASIAN JOURNAL OF CONTROL, 2011, 13 (06) : 903 - 913
[10] THE ERROR VARIANCE OF LMS WITH TIME-VARYING WEIGHTS
SOLO, V
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1992, 40 (04) : 803 - 813

← 1 2 3 4 5 →