請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/43069
標題: | 實現網頁瀏覽防護之檔案鑑別技術 Files Distinction Technology Achieving Protection of Web Page Browsing |
作者: | Chun-Ze Yang 楊淳澤 |
指導教授: | 郭斯彥 |
關鍵字: | 網頁安全,病毒蠕蟲,雲端偵測,可疑檔案判別, network security,malware,cloud computing,malicious files, |
出版年 : | 2009 |
學位: | 碩士 |
摘要: | 在網路發展熱潮的今日,也迫使網路犯罪的增加,導致網路病毒數量日益增加,而且技術越來越精進,經由瀏覽連結網頁,便可在網路頁開啟的同時植入病毒到使用者電腦,像病毒,蠕蟲,僵屍病毒,後門程式....等,隱藏在系統之中,達到遠端監控,個人資料竊取,這種行為往往不易被察覺,而藉由瀏覽連結網頁達成入侵的行為,成為近年來的主流,因此,為了預防網路上散播的病毒,有效防止誤連結惡意網站而遭受入侵,即時性的網頁偵測是很重要的。
本篇論文提出一個遠端處理架構,可以降低個人電腦負擔,並且讓使用者在連結網站時,立即判斷該網站是否藏有危機,讓用戶可以安全的保護下瀏覽網頁,而該架構面臨了反應時間的瓶頸,為了達成未來民生上的可行性,必需提升架構上的效能,針對此點,本論文提出了檔案鑑別技術,快速的將可疑的檔案過瀘出來,減少比對的次數和分析時間,以達到架構上理想狀態。 我們的技術是利用各種檔案格式的特徵碼,此方法可以將實際檔案型式判別出來,而刻意偽裝成一般檔案的病毒將無法隱暪,並且以幾個位元組就能判別檔案的去留,有利於提高檔案判別速度。 Internet services are rapidly developed today. They become essential part of our daily life. They not only force increase of network crime but also lead to increase number of malicious software in the network every day. The malicious software is also named as malware. The malware techniques are made more and more compact. When the webpage is turned on, the malware, like the viruses, the worms, the botnets, or the backdoor, are also planted to user’s computer and hided in the system via browsing the URL links at the same time. They spy on users’ computer and collect users’ private information by remote control. Users’ computers infected with malware cannot be aware of having been infected. So, the malware attack via browsing web pages is main way recently. Therefore, In order to prevent the malware invading from internet, real-time detection in the network is very important. This thesis proposes one structure of cloud computing to reduce the PC’s burden, and to help the user to judge immediately whether the linked websites hide crisis. The user can get safe protection of web page browsing. However, the structure of system has a bottleneck about responding time. We must improve efficiency in order to reach the feasibility on the people's livelihood in the future. For the purpose, we propose file distinction technology (FDT) in the paper. It can filter suspicious files fast and reduce analysis times for achieving the ideal state on the structure. Our technology is utilizing the signature of different file formats. This research can differentiate the real file formats from the fake ones. If the virus disguises as the general file sedulously, it will be unable to hide. This method can differentiate whether the files can be reserved with several bytes. It improves the speed of differentiating the file. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/43069 |
全文授權: | 有償授權 |
顯示於系所單位: | 電機工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-98-1.pdf 目前未授權公開取用 | 768.41 kB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。