|Table of Contents|

Comparison of Several Conducted EMI Soft-Separation Methods(PDF)

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

Issue:
2010年04期
Page:
4-7
Research Field:
Publishing date:

Info

Title:
Comparison of Several Conducted EMI Soft-Separation Methods
Author(s):
Ping Yunjuan; Qiu Xiaohui
School of Communication and Information,Nanjing University of Post and Telecommunication,Nanjing 210003,China
Keywords:
conductive EM I;  no ise sepa ra tion;  FastICA;  modern spectra l estim ation;  BP neura l ne tw ork
PACS:
TN03
DOI:
-
Abstract:
Fo r trad itional EM I no ise separa tion ne tw ork s’ poor h igh frequency cha racte ristics and expensive pr ice, th is paper com pares severa l signa l processing techniques, and finally propo ses a conducted EM I so ft sepa ra tion m e thod based on independen t com ponent analysis ( ICA ) techno logy, m odern spec trum estim a tion and BP neura l network a lgor ithm. Exper im enta l resu lt show s that the propo sed m ethod can rap idly and effective ly extract EM I conducted no isem odes, so as to prov ide a theo re tica l basis fo r EM I no ise suppression.

References:

[ 1] Paul C R, H ard in K B. D iagnosis and reduction o f conducted no ise em ission [ J]. IEEE Trans on EMC, 1988, 30 ( 4): 553-560.
[ 2] See K Y. Ne tw ork fo r conducted EM I d iagno sis[ J]. E lectron ic Letter, 1999, 35( 17): 1446-1447.
[ 3] M a rd iguianM, Raim bourg J. An a lternate comp lementarym ethod fo r characterizing EM I filters[ C] / / Proc of IEEE on EMC. Seattle: IEEE Press, 1999: 882-886.
[ 4] Guo T, Chen D Y. Separa tion o f the common-m ode and d ifferentia-l mode conducted EM I no ise[ J]. IEEE Trans Pow er E lectron ics, 1996, 11( 3): 480-488.
[ 5] Lo Y K, Ch iu H J, Song T H. A softw are-based CM and DM m easurem ent system for conducted EM I[ J]. IEEE T rans on Industrial E lectronics, 2000, 47( 4): 977-978.
[ 6] H yvarinen A, O ja E. A fast fixed-po int a lgo rithm for independent com ponent ana lysis[ J]. Neural com putation, 1997, 9( 7): 1483-1492.
[ 7] BrunoAudone. The use o f adaptive filters in the suppression of inter ference on signa l lines[ C] / / E lectrom agnetic Com pa tibil-i ty-EMC Europe, 2008 Inte rnationa l Symposium on. H am burg: IEEE Press, 2008.
[ 8] 吴丹宁. 现代信号处理在电磁干扰噪声分析中的应用[ D]. 南京: 南京邮电大学通信与信息工程学院, 2010: 23-24. W u Dann in. App lication of modern signa l processing technique in EM I no ise analysis[ D]. Nanjing: Schoo l of Comm un ication and Inform ation, N an jing University of Post and Telecomm un ica tion, 2010: 23-24. ( in Ch inese)
[ 9] M a llat S G. A theo ry form ultireso lution signa l decom pos ition: The w avelet representation[ J]. Patte rn Ana ly sis andM achine Inte lligence, IEEE Transactions on, 1989, 11( 7): 674-693.
[ 10] 姚天任, 孙洪. 现代数字信号处理[M ]. 武汉: 华中科技大学出版社, 1999. Yao T ianren, Sun H ong. Modern D ig ita l S ignal Pro cessing [M ]. Wuhan: H uazhong Un ive rsity o f Sc ience and Techno logy Press, 1999. ( in Ch inese)
[ 11] Q iu X iaohu,i W u Dann ing, Lu Xiaoquan, et a.l Identification o f conducted EM I no ise based on independent com ponent ana-l ysis and signa l sta tistics applied in underg round pow er e lectronic system [ J]. Pro ced ia Earth and Planetary Science, 2009, 1( 1): 1 581-1 586.
[ 12] 杨行峻, 郑君里. 人工神经网络[M ] . 北京: 高等教育出版社, 1992. Yang X ing jun, Zheng Jun l.i A rtific ia l Neura lNetw orks[M ]. Be ijing: H igher Education Press, 1992. ( in Chinese)

Memo

Memo:
-
Last Update: 2013-04-02