Presentation 2016-03-03
Throughput estimation using online machine learning algorithm from depth-images for mmWave communications
Hironao Okamoto, Takayuki Nishio, Masahiro Morikura, Koji Yamamoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In mmWave communications, received signal strength sharply decreases when pedestrians block line-of-sight (LOS) paths. To solve the human blockage problem, a base station handover has been assessed. In order to quickly transfer a mobile station communicating with one base station to another base station which provides higher throughput, the handover system obtains and updates throughput information between the station and each of base stations. Therefore, many control signals are required to transmit to obtain the throughput information and would waste mmWave radio resources. In this paper, we propose a mmWave throughput estimation scheme using an online machine learning algorithm and depth-image. A system with the scheme learns the relationships between depth-images and throughputs, and estimates throughputs from depth-images. The system enables optimized base station handover without control signals. The algorithm enables the handover system to estimate throughput fast and adaptively. We implemented a testbed using IEEE 802.11ad mmWave wireless LAN devices and an RGB-D camera. The experiment results confirm that the scheme estimates temporal variation of throughputs more accurately than a compared scheme in various scenarios.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) mmWave / machine learning / RGB-D camera / communication quality estimation / IEEE 802.11ad
Paper # SR2015-96
Date of Issue 2016-02-24 (SR)

Conference Information
Committee RCS / CCS / SR / SRW
Conference Date 2016/3/2(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Tokyo Institute of Technology
Topics (in Japanese) (See Japanese page)
Topics (in English) Mobile Communication Workshop
Chair Makoto Taromaru(Fukuoka Univ.) / Hiroo Sekiya(Chiba Univ.) / Takeo Fujii(Univ. of Electro-Comm.) / Hiroshi Harada(Kyoto Univ.)
Vice Chair Hidekazu Murata(Kyoto Univ.) / Satoshi Denno(Okayama Univ.) / Yukitoshi Sanada(Keio Univ.) / Yasuhiro Tsubo(Ritsumeikan Univ.) / Naoki Wakamiya(Osaka Univ.) / Kenta Umebayashi(Tokyo Univ. of Agric. and Tech.) / Masayuki Ariyoshi(NEC) / Masafumi Kato(Fujitsu) / Satoshi Denno(Okayama Univ.)
Secretary Hidekazu Murata(Mitsubishi Electric) / Satoshi Denno(NTT DoCoMo) / Yukitoshi Sanada(Kagawa National College of Tech.) / Yasuhiro Tsubo(Kyoto Sangyo Univ.) / Naoki Wakamiya(Shinshu Univ.) / Kenta Umebayashi(NICT) / Masayuki Ariyoshi(NTT) / Masafumi Kato(NICT) / Satoshi Denno
Assistant Jun Mashino(NTT) / Tetsuya Yamamoto(Panasonic) / Takamichi Inoue(NEC) / Tomoya Tandai(Toshiba) / Toshihiko Nishimura(Hokkaido Univ.) / Takayuki Kimura(Nippon Inst. of Tech.) / Song-Ju Kim(NIMS) / Ryo Takahashi(Kyoto Univ.) / Junnosuke Teramae(Osaka Univ.) / Kazuto Yano(ATR) / Mamiko Inamori(Tokai Univ.) / Hiroyuki Shiba(NTT) / Gia Khanh Tran(Tokyo Inst. of Tech.) / Wen Yun(Fujitsu) / Keiichi Mizutani(Kyoto Univ.)

Paper Information
Registration To Technical Committee on Radio Communication Systems / Technical Committee on Complex Communication Sciences / Technical Committee on Smart Radio / Technical Committee on Short Range Wireless Communications
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Throughput estimation using online machine learning algorithm from depth-images for mmWave communications
Sub Title (in English)
Keyword(1) mmWave
Keyword(2) machine learning
Keyword(3) RGB-D camera
Keyword(4) communication quality estimation
Keyword(5) IEEE 802.11ad
1st Author's Name Hironao Okamoto
1st Author's Affiliation Kyoto University(Kyoto Univ.)
2nd Author's Name Takayuki Nishio
2nd Author's Affiliation Kyoto University(Kyoto Univ.)
3rd Author's Name Masahiro Morikura
3rd Author's Affiliation Kyoto University(Kyoto Univ.)
4th Author's Name Koji Yamamoto
4th Author's Affiliation Kyoto University(Kyoto Univ.)
Date 2016-03-03
Paper # SR2015-96
Volume (vol) vol.115
Number (no) SR-473
Page pp.pp.47-52(SR),
#Pages 6
Date of Issue 2016-02-24 (SR)