Presentation | 2022-11-24 Study on Training Data Generation for Estimating Spatial Loss Fields Yoshiaki Nishikawa, Takahiro Matsuda, Eiji Takahashi, Takeo Onishi, Toshiki Takeuchi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Spatial Loss Fields (SLFs) are maps quantifying the attenuation of radio signals in a monitored region. SLFs, which are estimated from received signal strengths, have some kind of unclearness coming from reflections and diffractions of signal. Although surpervised learning methods to decrease this unclearness need plenty of true SLFs, it is dificult to collect true SLFs of various factories. It is needed to generate training data which are torelant to the multipath environment. In this article, we propose training data generation method with a simulation. The proposed method makes true SLF randomly, and simulates whether received signals are transmitted on a line-of-sight (LOS) path or a non-line-of-sight path. The model is trained to infer the difference between true and estimated SLFs from the estimated SLF. We use denoise convolutional neaural network as the model and evaluate the performance of the proposed method with simulation experience. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | spatial loss field / denoise convolutional neaural network / surpervised learning |
Paper # | CQ2022-47 |
Date of Issue | 2022-11-17 (CQ) |
Conference Information | |
Committee | NS / ICM / CQ |
---|---|
Conference Date | 2022/11/24(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Humanities and Social Sciences Center, Fukuoka Univ. + Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Network quality, Network measurement/management, Network virtualization, Network service, Blockchain, Security, Network intelligence/AI, etc. |
Chair | Tetsuya Oishi(NTT) / Yuji Nomura(Fujitsu) / Jun Okamoto(NTT) |
Vice Chair | Takumi Miyoshi(Shibaura Insti of Tech.) / Yu Miyoshi(NTT) / Eiji Takahashi(NEC) / Takefumi Hiraguri(Nippon Inst. of Tech.) / Gou Hasegawa(Tohoku Univ.) |
Secretary | Takumi Miyoshi(NTT) / Yu Miyoshi(Kogakuin Univ.) / Eiji Takahashi(NTT) / Takefumi Hiraguri(Fujitsu) / Gou Hasegawa(NTT) |
Assistant | Kotaro Mihara(NTT) / Ryo Yamamoto(Univ. of Electro-Comm) / Kimiko Kawashima(NTT) / Ryo Nakamura(Fukuoka Univ.) / Toshiro Nakahira(NTT) / Kenta Tsukatsune(Tokyo Metroplitan Univ.) |
Paper Information | |
Registration To | Technical Committee on Network Systems / Technical Committee on Information and Communication Management / Technical Committee on Communication Quality |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Study on Training Data Generation for Estimating Spatial Loss Fields |
Sub Title (in English) | |
Keyword(1) | spatial loss field |
Keyword(2) | denoise convolutional neaural network |
Keyword(3) | surpervised learning |
1st Author's Name | Yoshiaki Nishikawa |
1st Author's Affiliation | NEC(NEC) |
2nd Author's Name | Takahiro Matsuda |
2nd Author's Affiliation | Tokyo Metropolitan University(TMU) |
3rd Author's Name | Eiji Takahashi |
3rd Author's Affiliation | NEC(NEC) |
4th Author's Name | Takeo Onishi |
4th Author's Affiliation | NEC(NEC) |
5th Author's Name | Toshiki Takeuchi |
5th Author's Affiliation | NEC(NEC) |
Date | 2022-11-24 |
Paper # | CQ2022-47 |
Volume (vol) | vol.122 |
Number (no) | CQ-275 |
Page | pp.pp.1-6(CQ), |
#Pages | 6 |
Date of Issue | 2022-11-17 (CQ) |