A generalized normalized gradient descent algorithm

被引：195

作者：

Mandic, DP ^{[1
]}

机构：

[1] Univ London Imperial Coll Sci Technol & Med, Dept Elect & Elect Engn, London SW7 2BT, England

来源：

IEEE SIGNAL PROCESSING LETTERS | 2004年 / 11卷 / 02期

关键词：

adaptive filtering; gradient adaptive learning rate; nonlinear prediction; normalized least mean square;

D O I：

10.1109/LSP.2003.821649

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 generalized normalized gradient descent (CTNGD) algorithm for linear finite-impulse response (FIR) adaptive filters is introduced. The GNGD represents an extension of the normalized least mean square (NLMS) algorithm by means of an additional gradient adaptive term in the denominator of the learning rate of NLMS. This way, GNGD adapts its learning rate according to the dynamics of the input signal, with the additional adaptive term compensating for the simplifications in the derivation of NLMS. The performance of GNGD is bounded from below by the performance of the NLMS, whereas it converges in environments where NLMS diverges. The GNGD is shown to be robust to significant variations of initial values of its parameters. Simulations in the prediction setting support the analysis.

引用

页码：115 / 118

页数：4

共 11 条

[1] A robust variable step-size LMS-type algorithm: Analysis and simulations [J].

Aboulnasr, T ;

Mayyas, K .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1997, 45 (03) :631-639

[2] A new class of gradient adaptive step-size LMS algorithms [J].

Ang, WP ;

Farhang-Boroujeny, B .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2001, 49 (04) :805-810

[3]

Benveniste A, 1990, Adaptive algorithms and stochastic approximations

[4] ANALYSIS AND IMPLEMENTATION OF VARIABLE STEP-SIZE ADAPTIVE ALGORITHMS [J].

EVANS, JB ;

XUE, P ;

LIU, B .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1993, 41 (08) :2517-2535

[5]

KUZMINSKY AM, 1986, IZV VUZ RADIOELEKTR+, V29, P103

[6]

Mandic D, 2001, ADAPT LEARN SYST SIG, DOI 10.1002/047084535X

[7] A STOCHASTIC GRADIENT ADAPTIVE FILTER WITH GRADIENT ADAPTIVE STEP-SIZE [J].

MATHEWS, VJ ;

XIE, ZH .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1993, 41 (06) :2075-2087

[8]

Milisavljevic M, 2001, INT CONF ACOUST SPEE, P3853, DOI 10.1109/ICASSP.2001.940684

[9] On a class of computationally efficient, rapidly converging, generalized NLMS algorithms [J].

Morgan, DR ;

Kratzer, SG .

IEEE SIGNAL PROCESSING LETTERS, 1996, 3 (08) :245-247

[10]

Narendra K S, 1990, IEEE Trans Neural Netw, V1, P4, DOI 10.1109/72.80202

← 1 2 →