以虛實系統策略建構MSOC自動化防護機制：使用深度學習/通道狀態訊息與感測器/建築資訊模型技術;Development of Cyber-Physical System-Based Msoc Protection Mechanism: Using Deep Learning / Channel State Information and Sensors / Building Information Modeling

NCUIR > Engineering College > Department of Civil Engineering > Research Project > Item 987654321/79106

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/79106

Title:	以虛實系統策略建構MSOC自動化防護機制：使用深度學習/通道狀態訊息與感測器/建築資訊模型技術;Development of Cyber-Physical System-Based Msoc Protection Mechanism: Using Deep Learning / Channel State Information and Sensors / Building Information Modeling
Authors:	周建成
Contributors:	國立中央大學土木工程學系
Keywords:	建築資訊模型;深度學習;關鍵基礎設施相依性;通道狀態資訊;Building Information Modeling;Deep Learning;Critical Infrastructure Interdependency;Channel State Information.
Date:	2019-02-21
Issue Date:	2019-02-21 14:42:51 (UTC+8)
Publisher:	科技部
Abstract:	關鍵基礎設施相依性定義為重要設施或設備彼此的互相影響關係，若保護不當，可能有牽一髮動全身等連鎖反應。虛實系統定義為虛擬世界與真實世界的所有物件均可完整對應。國軍眾多設施/設備/機房，勢必有些設施極為重要，需要妥善保護否則影響遠大。本研究對於兩個以上的國軍重要設施，施以：(1)Wi-Fi SOC偵測，此為直接使用Wi-Fi訊號、非侵入式取得範圍內物件形狀之最新技術，將改寫應用至此；(2)機房耗電量偵測，基於團隊過去經驗將建立測試環境，量得機器與網路設備在駭客攻擊時之用電異常現象，發展深度學習演算法辨識；(3)其他感測器如聲音或動作感測器，在機房適當地點架設，並發展對應演算法。最後，將所有資安事件相關資料，併入建築資訊模型內，執行時空資料分析，讓管理者可知最新機房健康狀況。第二年著重真實環境與測試資料的建置，提升系統效能，最後期望對軍方重要設施的保護，有所提升並能擴展至更多機房加以保護。 ;Critical infrastructure interdependencies (CI) are defined as the interaction of important facilities or equipment. If not properly protected, there may be a chain reaction. Cyber-physical system (CPS) is defined as a virtual world and all objects in the real world can be completely corresponding. Many facilities/equipment/machine rooms of the National Army are bound to have some facilities that are extremely important and need to be properly protected or have a great impact. This study applies to two or more important national military facilities: (1) Wi-Fi SOC detection, which is the latest technology that directly uses Wi-Fi signals and non-intrusively obtains the shape of objects in the range, and will be rewritten to this research. (2) Power consumption detection in the equipment room, based on the team's past experience, will establish a test environment, measure the abnormality of power consumption of machines and network equipment during hacking attacks, and develop deep learning algorithm identification; (3) other senses The detector, such as a sound or motion sensor, is erected at the appropriate location in the computer room and the corresponding algorithm is developed. Finally, all the information related to the security incidents will be incorporated into the building information model, and the spatio-temporal data analysis will be performed to let the managers know the latest computer room health. In the second year, we focused on the establishment of real environment and test data to improve system performance. Finally, we hope that the protection of important military facilities will be improved and extended to more computer rooms for protection.
Relation:	財團法人國家實驗研究院科技政策研究與資訊中心
Appears in Collections:	[Department of Civil Engineering] Research Project

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	225	View/Open

社群 sharing

Loading...