Presentation 2020-07-09
Considerations on Accuracy Improvement in DOA Estimation Using Deep Learning
Yuya Kase, Takanori Sato, Toshihiko Nishimura, Takeo Ohgane, Yasutaka Ogawa, Daisuke Kitayama, Yoshihisa Kishiyama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Direction of arrival (DOA) estimation of radio waves using a various types of array antennas are generally classified into subspace methods such as MUSIC and ESPRIT, and probability distribution estimation such as EM and SAGE. Recently, compressed sensing and deep learning have been studied with a progress of computing resources. The compressed sensing and deep learning are on-grid estimation in general, and thus a discrete spectrum is obtained. In our previous studies on DOA estimation using deep learning, it was shown that the estimation frequently fails when a signal arrives at angles near the grid border. In this paper, we have proposed a method of combining two DNNs, of which grids are staggered, in order to reduce this estimation error. Simulation results show that the proposed combining method improves the estimation accuracy compared with the case where one DNN is used and achieves higher estimation accuracy than MUSIC.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) DOA estimation / array antenna / deep learning / deep neural network
Paper # RCC2020-5,NS2020-34,RCS2020-68,SR2020-13,SeMI2020-5
Date of Issue 2020-07-01 (RCC, NS, RCS, SR, SeMI)

Conference Information
Committee SR / NS / SeMI / RCC / RCS
Conference Date 2020/7/8(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Communication and Network Technology of the AI Age, M2M (Machine-to-Machine),D2D (Device-to-Device),IoT(Internet of Things), etc
Chair Masayuki Ariyoshi(NEC) / Akihiro Nakao(Univ. of Tokyo) / Susumu Ishihara(Shizuoka Univ.) / HUAN-BANG LI(NICT) / Eiji Okamoto(Nagoya Inst. of Tech.)
Vice Chair Suguru Kameda(Tohoku Univ.) / Osamu Takyu(Shinshu Univ.) / Kentaro Ishidu(NICT) / Tetsuya Oishi(NTT) / Kazuya Monden(Hitachi) / Koji Yamamoto(Kyoto Univ.) / Shunichi Azuma(Nagoya Univ.) / Koji Ishii(Kagawa Univ.) / Fumiaki Maehara(Waseda Univ.) / Toshihiko Nishimura(Hokkaido Univ.) / Tomoya Tandai(Toshiba)
Secretary Suguru Kameda(ATR) / Osamu Takyu(Univ. of Electro-Comm.) / Kentaro Ishidu(Mie Univ.) / Tetsuya Oishi(NTT) / Kazuya Monden(Chuo Univ.) / Koji Yamamoto(Kyoto Univ.) / Shunichi Azuma(Osaka Univ.) / Koji Ishii(Hitachi) / Fumiaki Maehara(Waseda Univ.) / Toshihiko Nishimura(CRIEPI) / Tomoya Tandai(Osaka Univ.)
Assistant Mai Ohta(Fukuoka Univ.) / Teppei Oyama(Fujitsu Lab.) / Kentaro Kobayashi(Nagoya Univ.) / Shinya Kawano(NTT) / Yuki Katsumata(NTT DOCOMO) / Yu Nakayama(Tokyo Univ. of Agri. and Tech.) / Akira Uchiyama(Osaka Univ.) / SHAN LIN(NICT) / Masaki Ogura(Osaka Univ.) / Koichi Adachi(Univ. of Electro-Comm.) / Osamu Nakamura(Sharp) / Manabu Sakai(Mitsubishi Electric) / Masashi Iwabuchi(NTT) / Tatsuki Okuyama(NTT DOCOMO)

Paper Information
Registration To Technical Committee on Smart Radio / Technical Committee on Network Systems / Technical Committee on Sensor Network and Mobile Intelligence / Technical Committee on Reliable Communication and Control / Technical Committee on Radio Communication Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Considerations on Accuracy Improvement in DOA Estimation Using Deep Learning
Sub Title (in English)
Keyword(1) DOA estimation
Keyword(2) array antenna
Keyword(3) deep learning
Keyword(4) deep neural network
1st Author's Name Yuya Kase
1st Author's Affiliation Hokkaido University(Hokkaido Univ.)
2nd Author's Name Takanori Sato
2nd Author's Affiliation Hokkaido University(Hokkaido Univ.)
3rd Author's Name Toshihiko Nishimura
3rd Author's Affiliation Hokkaido University(Hokkaido Univ.)
4th Author's Name Takeo Ohgane
4th Author's Affiliation Hokkaido University(Hokkaido Univ.)
5th Author's Name Yasutaka Ogawa
5th Author's Affiliation Hokkaido University(Hokkaido Univ.)
6th Author's Name Daisuke Kitayama
6th Author's Affiliation NTT DOCOMO, INC(NTT DOCOMO)
7th Author's Name Yoshihisa Kishiyama
7th Author's Affiliation NTT DOCOMO, INC(NTT DOCOMO)
Date 2020-07-09
Paper # RCC2020-5,NS2020-34,RCS2020-68,SR2020-13,SeMI2020-5
Volume (vol) vol.120
Number (no) RCC-87,NS-88,RCS-89,SR-90,SeMI-91
Page pp.pp.19-24(RCC), pp.19-24(NS), pp.61-66(RCS), pp.25-30(SR), pp.13-18(SeMI),
#Pages 6
Date of Issue 2020-07-01 (RCC, NS, RCS, SR, SeMI)