標題: 應用同形保密技術於社群適地性行動雲端服務 之研究
Preserving Privacy for Community Location-Based Mobile Cloud Applications Using Homomorphic Encryption
作者: 魯旺
W. K. Ruwan Indiaka Prasanna
王蒞君
Wang, Li Chun
電信工程研究所
關鍵字: 應用同形保密技術於社群適地性行動雲端服務 之研究;location based;homomorphic;secure;community;encryption
公開日期: 2010
摘要: 現今雲端運算 (Cloud Computing) 不僅只是基礎運算設施,同時也提供應用程式的服務。然而在雲端運算基礎架構下,資料儲存的位置已不同於傳統應用服務,大量且隱私的資料將可能在雲端運算架構的伺服端一覽無遺。因此,如何在雲端運算架構下增加資料安全性將是首要克服的問題。 本篇論文中,我們將討論在社群適地性行動雲端服務 (Community Location-based Mobile Cloud Services) 的資料安全性之研究。在傳統加密 (Traditional Encryption) 方法中最大的限制,即是資料加密後,所有資料處理都必須經過密碼文 (Cipher Text) 的解密 (Decryption) ,這樣的特性若在雲端運算環境下將可能造成隱私資料的外流。雲端運算架構最大的優點,在於提供使用者快速地資料運算,但同時使用者並不希望個人隱私資料被雲端伺服器所知道。我們提出基於XOR運算的同型保密技術 (Homomorphic Encryption Technique) 和分享位置機制,保證伺服端使用加密位置資料執行運算,使得使用者可以在雲端服務提供者中不透漏真實位置資料,分享自己的位置並獲得適地性服務 (Location-based Services) 。
Abstract Cloud computing provides computing infrastructures and applications as a service. Since the infrastructure of cloud computing is hosted by a third party, security has become a major concern. In this thesis we discuss the security issues of hosting servers for a location based social network service in the cloud computing infrastructure. The limitation of traditional encryption is that once data is encrypted, other processing cannot be done without decrypting the cipher text. This results in a con ict of interest in cloud computing environments. Users require cloud computing infrastructures to perform data processing, but they do not trust the cloud computing infrastructure with their sensitive information. A location sharing protocol, which permits server side calculations on encrypted location information, is proposed in this thesis. Based on the XOR homomorphic encryption technique, the proposed scheme allows the users to send their location as the cipher and obtain location based services without revealing location information to the cloud computing service providers.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079613621
http://hdl.handle.net/11536/42060
Appears in Collections:Thesis


Files in This Item:

  1. 362101.pdf
  2. 36210.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.