|Table of Contents|

The Target Profile Identification of Step Frequency MMW Radar Based on Wavelet Neural Network(PDF)

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
2002年01期
Page:
20-23
Research Field:
Publishing date:
2002-02-28

Info

Title:
The Target Profile Identification of Step Frequency MMW Radar Based on Wavelet Neural Network
Author(s):
LiYuehua ShenQinghong GaoDuntang LiXingguo ①
Department of Electronic Science and Technology, Nanjing University, Nanjing 210093)
Keywords:
signal processing radar targ ets image processing neural network w avelet transform recognit ion
PACS:
TN957.52
DOI:
-
Abstract:
By integrating w avelet t ransform w ith feed forw ard neural netw ork, a new adapt ive w avelet funct ion neural netw ork is proposed. The good localizat ion characteristics of w avelet funct ions in both time and f requency space allow hierarchical mult-i resolution learning of input-output data mapping . The wavelet shapes are adaptively computed to minimize an energy function for a specific applicat ion of radar targets. T he mathemat ical frame of the neural network is int roduced and error back propag at ion algorithm is used. The procedure of using w avelet neural netw ork for identif icat ion is described in detail. Based on the target specif ic informat ion offered by the range prof iles of step frequency MMW radar target s, the w avelet neural network is applied to recognition of three kinds of pract ical radar target s. Experiment results indicate that the w avelet neural netw ork has faster convergence speed and hig her correct recognit ion rate.

References:

1 李跃华, 李兴国.MUSIC 法用于频率步进毫米波雷达目标回波信号分析.电子测量与仪器学报, 1999, 13( 2) : 1~ 5
2 崔锦泰. 小波分析导论. 西安: 西安交通大学版社, 1995
3 Giles C L. Dynamics recurrent neural netwo rks. Theor y and application. IEEE Trans Neur al Network,1996, 5( 2) : 153~ 156
4 Mallat S. A theory for mut iresolution signal decomposition: t he wavelet repr esentation. IEEE Trans PAMI, 1989, 11( 7) : 674~ 684
5 Suz H H. Neural netw ork adaptive wavelets for signal r epresentation and classification. Optical Engineering , 1992, 31( 9) : 1 907~ 1 919

Memo

Memo:
-
Last Update: 2002-02-28