携帯端末による適応型クラウドコンピューティングシステムの設計と評価

2018 年度 実績報告書

• 研究課題/領域番号: 18J10020
• 研究機関: 東北大学
• 研究代表者: Gama Rodrigues Tiago 東北大学, 情報科学研究科, 特別研究員(DC2)
• 研究期間 (年度): 2018-04-25 – 2020-03-31
• キーワード: edge cloud / cloud computing / artificial intelligence / machine learning / computer networks / computation offloading

研究実績の概要

I developed a protocol that is capable of minimizing the number of necessary cloud servers in the Mobile Edge Cloud without hurting the quality given to users (more efficient resource usage and lower cost of deployment). I developed a power saving scheme for cloud access networks (less consumed energy with a high-quality service). I created a network deployment technique for post-disaster scenarios (cheapest and most efficient way to deploy communication equipment to connect victims). I presented a couple of conference works on how to optimize Machine Learning and customize it to work better with Mobile Edge Cloud. I completed a survey of research using Machine Learning and Mobile Edge Cloud (lists all existing works and gives detailed information and instructions for future research).

現在までの達成度 (区分)

1: 当初の計画以上に進展している

理由

The research followed the scheduled plan and the expected achievements were reached as planned.

今後の研究の推進方策

Because of the potential of Machine Learning, I have decided to change my original research plan. For my last year of Doctor's Course and JSPS scholarship, I plan to use Machine Learning to solve the difficult problem of Edge Cloud Computing server placement. This is a complicated issue, especially with multiple users. Solving this problem will help using Edge Cloud Computing with 5G and Internet of Things arguably the most important network paradigms of the next generation. Finally, I plan to finish the year with my Doctor's Course thesis on how to use Machine Learning to solve the problems of Edge Cloud Computing, from dynamic scenarios to multiple users, finalizing with an overarching base on how to make this useful technology realistic and bring it to the public.

研究成果

• [雑誌論文] An Internet of Things Traffic-Based Power Saving Scheme in Cloud-Radio Access Network (2019)
  • 著者名/発表者名: Miyanabe Keisuke、Gama Rodrigues Tiago、Lee Yunseong、Nishiyama Hiroki、Kato Nei
  • 雑誌名: IEEE Internet of Things Journal 巻: 6 ページ: 3087～3096
  • DOI: 10.1109/JIOT.2018.2878884
  • 査読あり / 国際共著
• [雑誌論文] Cloudlets Activation Scheme for Scalable Mobile Edge Computing with Transmission Power Control and Virtual Machine Migration (2018)
  • 著者名/発表者名: Rodrigues Tiago Gama、Suto Katsuya、Nishiyama Hiroki、Kato Nei、Temma Katsuhiro
  • 雑誌名: IEEE Transactions on Computers 巻: 67 ページ: 1287～1300
  • DOI: 10.1109/TC.2018.2818144
  • 査読あり / 国際共著
• [雑誌論文] A Probabilistic Approach to Deploying Disaster Response Network (2018)
  • 著者名/発表者名: Nishiyama Hiroki、Rodrigues Tiago Gama、Liu Jiajia
  • 雑誌名: IEEE Transactions on Vehicular Technology 巻: 67 ページ: 12086～12094
  • DOI: 10.1109/TVT.2018.2872542
  • 査読あり / 国際共著
• [学会発表] Cross-Acceleration Study of Particle Swarm Optimization Applied to Resource Minimization in Mobile Edge Computing (2018)
  • 著者名/発表者名: Gama Rodrigues Tiago
  • 学会等名: IEICE