NONLINEAR PARAMETER-ESTIMATION VIA THE GENETIC ALGORITHM

被引：161

作者：

YAO, L ^{[1
]}

SETHARES, WA ^{[1
]}

机构：

[1] UNIV WISCONSIN,DEPT COMP ENGN,MADISON,WI 53706

来源：

IEEE TRANSACTIONS ON SIGNAL PROCESSING | 1994年 / 42卷 / 04期

关键词：

Algorithms - Digital filters - Networks (circuits) - Neural networks - Parameter estimation;

D O I：

10.1109/78.285655

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：

A modified genetic algorithm is used to solve the parameter identification problem for linear and nonlinear digital filters. Under suitable hypotheses, the estimation error is shown to converge in probability to zero. The scheme is also applied to feedforward and recurrent neural networks.

引用

页码：927 / 935

页数：9

共 27 条

[1]

[Anonymous], 1987, GENETIC ALGORITHMS S

[2]

ASH RB, 1970, BASIC PROBABILITY TH

[3]

CHOWNING J, 1986, FM THEORY APPLICATIO

[4]

CHOWNING JM, 1973, J AUDIO ENG SOC, V21, P526

[5] DISTRIBUTED GENETIC ALGORITHMS FOR THE FLOORPLAN DESIGN PROBLEM [J].

COHOON, JP ;

HEGDE, SU ;

MARTIN, WN ;

RICHARDS, DS .

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 1991, 10 (04) :483-492

[6]

Davidor Y., 1991, GENETIC ALGORITHMS R, V1

[7]

DAVIS L, 1987, GENETIC ALGORITHMS S, P1

[8]

De Jong K., 1988, Machine Learning, V3, P121, DOI 10.1023/A:1022606120092

[9]

DE JONG K. A., 1985, P INT C GEN ALG THEI, P169

[10]

DEJONG KA, 1988, 5TH P INT MACH LEARN, P284

← 1 2 3 →