THE OPTIMUM SCALAR DATA NONLINEARITY IN LMS ADAPTATION FOR ARBITRARY IID INPUTS

被引：7

作者：

DOUGLAS, SC

MENG, THY

机构：

[1] Information Systems Laboratory, Stanford University, Stanford

来源：

IEEE TRANSACTIONS ON SIGNAL PROCESSING | 1992年 / 40卷 / 06期

关键词：

D O I：

10.1109/78.139261

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 this correspondence, we show that the optimum nonlinear scalar operation upon the elements of the observation vector in the LMS algorithm is exactly x/(1 + mu-x2) for any independent stochastic data input and any noise density. Moreover, use of such a nonlinearity can yield a significant performance improvement in fast adaptation situations.

引用

页码：1566 / 1570

页数：5

共 8 条

[1] ON THE OPTIMUM DATA NONLINEARITY IN LMS ADAPTATION [J].

BERSHAD, NJ .

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1986, 34 (01) :69-76

[2]

COURANT R, 1953, METHODS MATH PHYSICS, V1, P164

[3]

DOUGLAS SC, THESIS STANFORD U

[4] ADAPTIVE FILTER PERFORMANCE WITH NONLINEARITIES IN THE CORRELATION MULTIPLIER [J].

DUTTWEILER, DL .

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1982, 30 (04) :578-586

[5] LEARNING CHARACTERISTICS OF STOCHASTIC-GRADIENT-DESCENT ALGORITHMS - A GENERAL STUDY, ANALYSIS, AND CRITIQUE [J].

GARDNER, WA .

SIGNAL PROCESSING, 1984, 6 (02) :113-133

[6] DIGITAL SUBSCRIBER LINES - NETWORK CONSIDERATIONS FOR ISDN BASIC ACCESS STANDARD [J].

HUANG, DT ;

VALENTI, CF .

PROCEEDINGS OF THE IEEE, 1991, 79 (02) :125-145

[7]

SETHARES WA, 1989, IEEE T ACOUST SPEECH, V38, P138

[8]

Widrow B., 1985, ADAPTIVE SIGNAL PROC

← 1 →