|Table of Contents|

New Radio Fuze Target Recognition Based on Wavelet Neural Network(PDF)

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2010年02期
Page:
84-87
Research Field:
Publishing date:

Info

Title:
New Radio Fuze Target Recognition Based on Wavelet Neural Network
Author(s):
Shan Jianfeng1; Zhai Bo2
1.College of Electronic Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;2.School of Computer and Communication Engineering,Liaoning Shihua University,Fushun 113001,China
Keywords:
w ave let transfo rm (WT);  rad io fuze;  feature extrac tion;  neura l ne tw ork
PACS:
TJ430
DOI:
-
Abstract:
New rad io fuze target recogn ition based on w av elet neura l netwo rk is d iscussed in th is paper. The radio fuze signal is first decomposed by wave let transform, and on this basis, the decomposed coeffic icien ts are reconstructed to form a new tim e ser ies, from w hich som e energy param eters can be ex tracted by tim e-dom a in ana lys is. The targe t signal is detected by w ave let neural ne tw ork. The e ffectiv eness o f them ethod is ve rified by a typ ica l radio fuze signals w ith d i-f ferent S igna-l to-No ise Ratio( SNR). The exper im enta l resu lts show tha t SNR processed by them ethod can reach - 7 dB.

References:

[ 1] E tem adiA H, Sanaye-PasandM. H igh- im pedance fau lt detec tion using mu lt-i reso lution s igna l decomposition and adaptive neura l fuzzy infe rence system [ J]. IET Gener Transm D istr ib, 2008, 2( 1): 110-118.
[ 2] Serhat Seker, Em ine Ayaz. Fea ture extraction re lated to bear ing dam age in e lec tricm o to rs by w avelet analysis[ J]. Journa l o f the Franklin Institu te, 2003, 340( 2): 125-134.
[ 3] KhanM A S K, Azizur RahmanM. Developm ent and imp lem entation o f a nove l fault d iagnostic and pro tection techn ique for IPM m o to r dr ives[ J]. IEEE T ransactions on Industr ia l E lectron ics, 2009, 56( 1) : 85-92.
[ 4] Ioann is P Tsoum as, Geo rge Georgou las, Epam inondas D M itron ikas, et a.l Asynchronous m ach ine rotor fau lt d iagnos is technique using comp lex w ave lets[ J] . IEEE Transactions on Ene rgy Convers ion, 2008, 23( 2) : 444-459.
[ 5] Barkat B, Boashash B. A high- reso lution quadratic tim e- frequency d istribution for mu lticom ponent signals ana lysis[ J]. IEEE Trans on Signa l Pro cessing, 2002, 49: 2 232-2 239.
[ 6] Inak i Rom ero, Ne il R. T-w ave a lternans found in preventr icular tachyarrhythm ias in CCU patients us ing a w ave let transformbased m ethodo logy [ J] . IEEE T ransactions on B iom ed ica l Eng ineer ing, 2008, 55( 11): 2 658-2 665.
[ 7] W ang Yong jin, Guan L ing. Recogn izing hum an emo tiona l sta te from aud iov isua l s igna ls[ J]. IEEE T ransactions onM ultim ed-i a, 2008, 10( 4): 659-668.
[ 8] Avc i E. Perform ance comparison o f w ave let fam ilies for analog m odulation c lassification us ing expe rt d iscre te w avelet neural network system [ J]. Expe rt System sW ith Applications, 2007, 33( 1) : 23-35.
[ 9] Avci E, Turkog lu I, PoyrazM. Inte lligent target recogn ition based on w avelet packe t neura l ne tw ork[ J]. Experts System sW ith App lica tions, 2005, 29( 1): 175-182.
[ 10] Avci E, Avc iD. The pe rfo rm ance compar ison of d iscre tew avelet neural ne tw ork and discrete wave le t adaptive netw ork based fuzzy inference system for d ig ita lm odulation recogn ition[ J]. Expert System sW ith Applications, 2007, 35( 1): 90-101.

Memo

Memo:
-
Last Update: 2013-04-02