| Presentation | 2022-09-14 A Study of Building Map Representation for Spatiotemporal Channel Parameters Estimation Model by Machine Learning Keiji Yoshikawa, Tatsuya Nagao, Kazuki Takezawa, Takahiro Hayashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Wireless emulators are being developed to design and evaluate wireless systems in virtual space. To emulate various environments with high accuracy, a highly accurate propagation model is required for individual environments. In particular, the modeling of not only propagation loss but also spatiotemporal characteristics is necessary to verify fading variations due to multiple paths. Recently, machine learning methods using spherical images and building maps have been proposed as models for estimating site-specific propagation characteristics. However, information on buildings where reflections and diffractions on multiple paths occur is important for modeling spatiotemporal characteristics. These reflections and diffractions occur at various locations out of sight of the transmitter and receiver and occur in large numbers in the vicinity of the transmitter and receiver. Therefore, depending on the representation format of the buildings, there is a concern that accuracy may be degraded due to insufficient information, and the data size may be increased due to unnecessary information. This paper proposes an input format that represents buildings based on polar coordinates centered on the transmitter/receiver for estimation that takes buildings into account. The effectiveness of the proposed method is verified through a simulation evaluation in which the delay spread is calculated as a spatiotemporal parameter. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | radio propagation prediction / spatiotemporal parameters / machine learning / building map |
| Paper # | AP2022-78 |
| Date of Issue | 2022-09-07 (AP) |
| Conference Information | |
| Committee | AP / MW |
|---|---|
| Conference Date | 2022/9/14(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | The Museum of Art, EHIME |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Microwave, Millimeter wave |
| Chair | Hiroshi Yamada(Niigata Univ.) / Noriharu Suematsu(Tohoku Univ.) |
| Vice Chair | Mitoshi Fujimoto(Fukui Univ) / Tadashi Kawai(Univ. of Hyogo) / Kensuke Okubo(Okayama Prefectural Univ.) / Hideyuki Nakamizo(Mitsubishi Electric) |
| Secretary | Mitoshi Fujimoto(National Defense Academy) / Tadashi Kawai(Mitsubishi Electric) / Kensuke Okubo(Univ. of Electro-Comm) / Hideyuki Nakamizo(Toshiba) |
| Assistant | Tomoki Murakami(NTT) / Naoki Hasegawa(Softbank) / Kosuke Katayama(NIT Tokuyama College) |
| Paper Information | |
| Registration To | Technical Committee on Antennas and Propagation / Technical Committee on Microwaves |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Study of Building Map Representation for Spatiotemporal Channel Parameters Estimation Model by Machine Learning |
| Sub Title (in English) | |
| Keyword(1) | radio propagation prediction |
| Keyword(2) | spatiotemporal parameters |
| Keyword(3) | machine learning |
| Keyword(4) | building map |
| 1st Author's Name | Keiji Yoshikawa |
| 1st Author's Affiliation | KDDI Research, Inc(KDDI Research, Inc) |
| 2nd Author's Name | Tatsuya Nagao |
| 2nd Author's Affiliation | KDDI Research, Inc(KDDI Research, Inc) |
| 3rd Author's Name | Kazuki Takezawa |
| 3rd Author's Affiliation | KDDI Research, Inc(KDDI Research, Inc) |
| 4th Author's Name | Takahiro Hayashi |
| 4th Author's Affiliation | KDDI Research, Inc(KDDI Research, Inc) |
| Date | 2022-09-14 |
| Paper # | AP2022-78 |
| Volume (vol) | vol.122 |
| Number (no) | AP-182 |
| Page | pp.pp.38-43(AP), |
| #Pages | 6 |
| Date of Issue | 2022-09-07 (AP) |