Place-and-Play Complex-Event Processing Engine on Dynamic Ad-hoc Networks

• Item type: Symposium(1)
• 公開日: 2017-06-21
• タイトル: Place-and-Play Complex-Event Processing Engine on Dynamic Ad-hoc Networks
• タイトル / 言語: en
• タイトル: Place-and-Play Complex-Event Processing Engine on Dynamic Ad-hoc Networks
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: Edge Computing,Distributed Processing,Complex Event Processing,Stream Processing
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_5794
• 資源タイプ: conference paper
• 著者所属: Osaka University
• 著者所属: Osaka University
• 著者所属: Osaka University
• 著者所属: KDDI
• 著者所属: KDDI
• 著者所属(英): en
• 著者所属(英): Osaka University
• 著者所属(英): en
• 著者所属(英): Osaka University
• 著者所属(英): en
• 著者所属(英): Osaka University
• 著者所属(英): en
• 著者所属(英): KDDI
• 著者所属(英): en
• 著者所属(英): KDDI
• 著者名: Sunyanan, Choochotkaew; Hirozumi, Yamaguchi; Teruo, Higashino; Megumi, Shibuya; Teruyuki, Hasegawa
• 著者名(英): Sunyanan, Choochotkaew; Hirozumi, Yamaguchi; Teruo, Higashino; Megumi, Shibuya; Teruyuki, Hasegawa
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In this paper, we introduce a general complex event processing (CEP) engine aiming for accomplishing at dynamic ad-hoc sensor networks with place-and-play concept (less configuration). In the perspective of the processing tool, we introduce a pseudo-source mechanism to cover wider application ranges and to obsolete prerequisites of source specification in conventional stream processing, and we also define a brand-new event specification language to support our developed system. To prevent data overflow and privacy issues and draw full power of the edge devices, our proposed system has been designed to operate on behind-edge networks which are broadly considered as dynamic ad-hoc networks with limited resources. Our system integrates a resource-constraint optimization and forest-like dependence grouping algorithm to autonomously distribute multiple processing tasks over the networks in a consistent and efficient way considering total flow on the path from sources to any destination placed in the networks. A large-scale simulation with a realistic smart-building scenario shows that our proposed method achieves about 6.6 times smaller flow volume and 2 times lower loss rate compared to centralization and is relatively superior to hop-based distribution approach. Notably, a prototype engine is successfully deployed over an ad-hoc wireless sensor and actuator network through Intel Edison modules in the real environment.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In this paper, we introduce a general complex event processing (CEP) engine aiming for accomplishing at dynamic ad-hoc sensor networks with place-and-play concept (less configuration). In the perspective of the processing tool, we introduce a pseudo-source mechanism to cover wider application ranges and to obsolete prerequisites of source specification in conventional stream processing, and we also define a brand-new event specification language to support our developed system. To prevent data overflow and privacy issues and draw full power of the edge devices, our proposed system has been designed to operate on behind-edge networks which are broadly considered as dynamic ad-hoc networks with limited resources. Our system integrates a resource-constraint optimization and forest-like dependence grouping algorithm to autonomously distribute multiple processing tasks over the networks in a consistent and efficient way considering total flow on the path from sources to any destination placed in the networks. A large-scale simulation with a realistic smart-building scenario shows that our proposed method achieves about 6.6 times smaller flow volume and 2 times lower loss rate compared to centralization and is relatively superior to hop-based distribution approach. Notably, a prototype engine is successfully deployed over an ad-hoc wireless sensor and actuator network through Intel Edison modules in the real environment.
• 書誌情報: マルチメディア，分散協調とモバイルシンポジウム2017論文集; 巻 2017, p. 538-549, 発行日 2017-06-21
• 出版者 / 言語: ja
• 出版者: 情報処理学会