Presentation 2022-10-13
An estimation of magnetic coupling coefficient between parallel two MSLs using machine leaning of near field information
Yusuke Sato, Sho Muroga, Hidefumi Kamozawa, Motoshi Tanaka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) An estimation method of the magnetic coupling coefficient between printed-circuit-board-level lines using a near field information was investigated. Parallel two microstrip lines (MSLs) with different distances between the lines were used as a test bench. A signal current of 1 GHz was applied to one MSL. Both ends of the other MSL were terminated with 50 Ω. The current composed of the signal current in one MSL and the return current in the ground and the neighboring MSL was modeled as a simple one-turn equivalent loop current model. First, a classifier was developed by using a one-dimensional CNN (Convolutional neural network) learned with the theoretical values of the near-field distribution generated from the loop current model. Next, the simulated near-field distributions above the parallel two MSLs were input to the classifier to estimate the geometry of the loop current models. The magnetic coupling coefficients between the estimated loop current models were calculated and compared with those between the parallel two MSLs.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Magnetic coupling coefficient / Magnetic near-field / Loop current model / Microstrip line / Machine learning / Convolutional neural network
Paper # EMCJ2022-35,MW2022-81,EST2022-45
Date of Issue 2022-10-06 (EMCJ, MW, EST)

Conference Information
Committee EMCJ / MW / EST / IEE-EMC
Conference Date 2022/10/13(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Akita University
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Atsuhiro Nishikata(Tokyo Inst. of Tech.) / Noriharu Suematsu(Tohoku Univ.) / Jun Shibayama(Hosei Univ.)
Vice Chair Kimihiro Tajima(NTT-AT) / Tadashi Kawai(Univ. of Hyogo) / Kensuke Okubo(Okayama Prefectural Univ.) / Hideyuki Nakamizo(Mitsubishi Electric) / Masayuki Kimishima(Advantest) / Yasuhide Tsuji(Muroran Inst. of Tech.) / Yasuo Ohtera(Toyama Prefectural Univ.)
Secretary Kimihiro Tajima(Hokkaido Univ.) / Tadashi Kawai(Hitachi) / Kensuke Okubo(Univ. of Electro-Comm) / Hideyuki Nakamizo(Toshiba) / Masayuki Kimishima(yoto Univ.) / Yasuhide Tsuji(Univ. of Electro-Comm) / Yasuo Ohtera
Assistant Kiyoto Matsushima(Hitachi) / Kenji Ogata(ADOX) / Toru Matsushima(Kyushu Inst. of Tech.) / Naoki Hasegawa(Softbank) / Kosuke Katayama(NIT Tokuyama College) / Seiya Kishimoto(Nihon Univ.) / Akito Iguchi(Muroran Inst. of Tech)

Paper Information
Registration To Technical Committee on Electromagnetic Compatibility / Technical Committee on Microwaves / Technical Committee on Electronics Simulation Technology / Technical Meeting on Electromagnetic Compatibility
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An estimation of magnetic coupling coefficient between parallel two MSLs using machine leaning of near field information
Sub Title (in English)
Keyword(1) Magnetic coupling coefficient
Keyword(2) Magnetic near-field
Keyword(3) Loop current model
Keyword(4) Microstrip line
Keyword(5) Machine learning
Keyword(6) Convolutional neural network
1st Author's Name Yusuke Sato
1st Author's Affiliation Akita University(Akita Univ.)
2nd Author's Name Sho Muroga
2nd Author's Affiliation Akita University(Akita Univ.)
3rd Author's Name Hidefumi Kamozawa
3rd Author's Affiliation Akita University(Akita Univ.)
4th Author's Name Motoshi Tanaka
4th Author's Affiliation Akita University(Akita Univ.)
Date 2022-10-13
Paper # EMCJ2022-35,MW2022-81,EST2022-45
Volume (vol) vol.122
Number (no) EMCJ-206,MW-207,EST-208
Page pp.pp.1-5(EMCJ), pp.1-5(MW), pp.1-5(EST),
#Pages 5
Date of Issue 2022-10-06 (EMCJ, MW, EST)