Presentation | 2023-05-19 An experimental evaluation of millimeter-wave link quality prediction using Wi-Fi CSI and supervised learning Shoki Ohta, Kanare Kodera, Takayuki Nishio, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This study experimentally evaluates our 60 GHz band millimeter-wave (mmWave) link quality prediction method using 5 GHz band Wi-Fi channel state information (CSI) acquired at multiple locations and machine learning. Wireless communication using high-frequency waves such as mmWaves enables a high transmission rate due to its wide bandwidth. However, mmWave link quality significantly deteriorates when the mmWave line-of-sight (LOS) path is blocked by obstacles such as humans or vehicles. Existing studies have proposed methods that use computer vision information such as images or point clouds and machine learning to predict blockage, but there are many environments where acquiring images or point clouds is difficult due to privacy or cost concerns. In this study, we propose a method using 5 GHz band Wi-Fi CSI and supervised learning to predict the degradation of link quality due to LOS blockage. 5 GHz band CSI contains less privacy information compared to images or point clouds, and measurement devices are inexpensive and easy to install. However, CSI has difficulty understanding information about the position and movement of objects that block the mmWaveLOS path compared to computer vision information such as images and point clouds. We enabled a detailed understanding of mmWave propagation space information by installing CSI measurement devices in multiple locations. Experiment evaluations in indoor environments demonstrated that 5 GHz Wi-Fi CSI acquired from multiple locations can predict significant degradation of mmWave communication throughput 500 ms ahead. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | link quality prediction / millimeter-wave communication / CSI / machine learning / Wi-Fi |
Paper # | SeMI2023-10 |
Date of Issue | 2023-05-11 (SeMI) |
Conference Information | |
Committee | SeMI / IPSJ-ITS / IPSJ-MBL / IPSJ-DPS |
---|---|
Conference Date | 2023/5/18(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Okinawa Institute of Science and Technology (OIST) |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Koji Yamamoto(Kyoto Univ.) / Yuichi Tokunaga(Kanazawa Institute of Technology) / Hirozumi Yamaguchi(Osaka University) / Takuo Suganuma(Tohoku University) |
Vice Chair | Kazuya Monden(Hitachi) / Yasunori Owada(NICT) / Shunsuke Saruwatari(Osaka Univ.) |
Secretary | Kazuya Monden(NTT DOCOMO) / Yasunori Owada(Tokyo Univ. of Agri. and Tech.) / Shunsuke Saruwatari(Osaka Univ.) |
Assistant | Yuki Matsuda(NAIST) / Akihito Taya(Aoyama Gakuin Univ.) / Takeshi Hirai(Osaka Univ.) |
Paper Information | |
Registration To | Technical Committee on Sensor Network and Mobile Intelligence / Special Interest Group on Intelligent Transport Systems and Smart Community / Special Interest Group on Mobile Computing and Smart Society System / Special Interest Group on Distributed Processing System |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | An experimental evaluation of millimeter-wave link quality prediction using Wi-Fi CSI and supervised learning |
Sub Title (in English) | |
Keyword(1) | link quality prediction |
Keyword(2) | millimeter-wave communication |
Keyword(3) | CSI |
Keyword(4) | machine learning |
Keyword(5) | Wi-Fi |
1st Author's Name | Shoki Ohta |
1st Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
2nd Author's Name | Kanare Kodera |
2nd Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
3rd Author's Name | Takayuki Nishio |
3rd Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
Date | 2023-05-19 |
Paper # | SeMI2023-10 |
Volume (vol) | vol.123 |
Number (no) | SeMI-31 |
Page | pp.pp.42-45(SeMI), |
#Pages | 4 |
Date of Issue | 2023-05-11 (SeMI) |