| Authors |
Y. Iyoda, K. Kobayashi, H. Okada, Chedlia Ben Naila, M. Katayama |
| Title |
A study on robustness against symbol misalignment and inter-symbol interference on a CNN-based demodulation method for image sensor-based visible light communication |
| Authority |
International Conference on Materials and Systems for Sustainability (ICMaSS), A6-III-4 |
| Summary |
This study focuses on the demodulation part of the communication system, which modulates data signals without interfering with the visual information displayed on the digital signage.
We have studied a method to demodulate the transmitted data signals from the received image of the image sensor using machine learning with data sets that simulate noise, blur, and misalignment that may occur in the received images.
In this paper, we propose to train the demodulation method on datasets with misalignment of the data symbols to be demodulated and inter-symbol interference from surrounding symbols on the simulated received images to acquire robustness against symbol misalignment and inter-symbol interference.
We also evaluate the obtained robustness of the machine learning model.
|
| 年月 |
2023年12月 |
| DOI/Handle |
|
| 開催場所 |
Nagoya, Japan |
| 研究テーマ |
可視光通信／光無線通信
機械学習
|
| 言語 |
英語 |
| 原稿/プレゼン資料 |
 /  (ローカル限定) |
