| Project/Area Number |
17K00136
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Information network
|
|---|
| Research Institution | Tokyo University of Science |
|---|
Principal Investigator |
Hasegawa Mikio 東京理科大学, 工学部電気工学科, 教授 (50358967)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2018: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2017: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
|---|
| Keywords | 無線通信システム / 機械学習 / Multi-Armed Bandit問題 / 強化学習 / IoT / センサネットワーク / 無線LAN / Mult-Armed Bandit問題 / ニューラルネットワーク / サポートベクトルマシン / コグニティブ無線 / ディープラーニング / 情報通信工学 / 移動体通信 |
|---|
| Outline of Final Research Achievements |
In order to improve the performance of distributed wireless networks without centralized control, this research has proposed a machine-learning-based distributed decision making method for each terminal to optimize entire networks. Effectiveness of the proposed scheme has been clarified not only by computer simulations but also by implementation and experiments. For IoT, we have proposed a light-weight reinforcement learning algorithm, and have shown effectiveness of the proposed algorithms by implementing the method on IoT devices and experiments in the real wireless environment. Our contribution on AI-based algorithm implemented on the IoT devices has been featured in various scientific news sites [ex. QS WOWNEWS, Science Daily, Tech Explorist, Technology Networks, IoTToday, @IT. NIKKEI BUSINESS DAILY]. We have shown the possibility of improvement of the network performance by distributed machine learning, and we expect its application to various wireless communication systems.
|
| Academic Significance and Societal Importance of the Research Achievements |
無線ネットワークのつながりやすさや通信速度を改善するために，無線機器に搭載する人工知能によって，様々な通信パラメータを最適化する研究である．実際の無線通信を用いた実機実験によって提案手法の有効性を実証してきた．特に，近年普及が進んでいるIoT向けの無線通信システムを用いた提案手法およびその実証は，国内外の多くのニュースサイト等に掲載された（日経産業新聞、QS WOWNEWS、Science Daily、Tech Explorist、Technology Networks、IoTToday、@ITなど）．既に複数の無線通信システムで有効性実証ができており，様々なシステムへの応用が期待できる．
|
