- 學位論文
在雲端服務提供具使用者詢答隱私保護之資料搜尋功能
Privacy Preservation for User Queries and Search Results in Cloud Services
摘要
近幾年雲端服務相當興盛，並且大幅改變人們使用網際網路的方式。許多軟體服務商透過雲端帶來的資源，提供更完善的服務品質與內容給他們的用戶。在各類形式的應用服務中，資料詢答相關功能往往佔有重要的一席之地。然而，資料詢答過程中，若沒有特別進行隱私防護措施，使用者進行詢答過程，多數情況不知其詢問內容已對服務提供者或有心人士透露出隱私資訊，如使用者關心的資訊、個人重要資料以及重要檔案內容等。若讓有意圖的人士取得，則可能利用這些資訊對使用者進行非預期甚至不法的行為。在維護使用者隱私以及讓詢答資訊的搜尋功能在雲端上應用，這兩者在目的與需求上，看起來似乎是有些衝突的。 對於使用者來說，詢答情境主要可分為兩種。一種是待詢答資料集合屬於服務提供者可解讀的，如搜尋引擎之於其所蒐集的資料；另一種則是不能直接解讀的，如雲端儲存服務之於使用者加密儲存於其中的資料。若要保護使用者隱私，對於前者來說，必須隱藏其詢問資訊，並且從詢答結果不容易知道使用者想詢問的資訊；而對於後者來說，若要提供詢答機制，除了詢問資訊需要保護外，使用者亦需提供雲端服務提供者可搜尋與維護的索引結構。 針對上述兩類情況，首先在我們的研究中，以提出可用來建置具使用者隱私保護的雲端資安服務系統框架來探討第一種詢答情境。以資安服務來說，對於用戶裝置在資源有所限制的情況，資安服務商除了在用戶裝置提供基礎的資安分析功能外，更透過以詢答為基礎的方式，結合雲端資安服務所擁有之不斷精進的智能，協助用戶有效抵抗網路的攻擊。透過雲端大量的硬體資源支援以及資安服務商的專業分析，以雲端服務為基礎的資安服務比起傳統的主機型資安監控，可以帶來更完善的防護。此系統框架涵蓋可以用來建立雲端服務的架構，並具備私人特徵過濾技術，用以保護使用者隱私。而這個系統框架所擁有的資安監控特徵值可以用來分析比對來自用戶裝置，透過關鍵字與數字範圍聯集所組成的詢問。此外，在此框架之中，也可利用一個信任的雲端中介層，進行部分用戶裝置所需要的資安監控運算，減少用戶裝置在資源方面的消耗。在我們的研究中，實作了雲端網路入侵偵測與雲端惡意網址偵測雛型系統，用以確定我們所設計的框架之有效性。由實驗結果可知，我們所設計的系統框架在實際雲端資安服務的設置情況下，的確能夠在可接受的效能表現下確保使用者的隱私。 另外一方面，雲端儲存服務對於一般大眾所帶來的便利性以及資料保存性等優點，使其受歡迎的程度不斷升高。用戶只需將其資料存放在雲端儲存服務商所提供的雲端空間，就可以任何裝置，在任何時間與地點，透過網際網路存取資料。而以資料安全考量來說，個人資料存放在雲端時，必須透過資料加密的方式來避免敏感性資料洩漏給不信任的第三方單位。然而隨著存放在雲端的資料越來越多，要能夠快速且安全地搜尋到個人加密過的資料放在哪裡是非常困難的。 有鑑於此，對於第二種詢答情境，我們以雲端儲存服務為例，提出一個可以支援加密格式搜尋的系統框架，讓用戶可以快速且安全地搜尋到存放在雲端空間的資料。在系統框架的核心，是一個創新的索引結構，稱為布隆過濾器加密搜尋樹(BFEST, bloom filter encrypted search tree)，並且支援片語關鍵字的方式搜尋。透過使用者端的加密，並且雲端服務提供者不知道實際的密鑰為何，因此在資料取得以及搜尋內容方面，均可有效保護。為了確定研究成果如預期，我們在hicloud S3雲端儲存服務實作此研究設計的雛型機制。而由實驗結果可知，此系統框架同樣可以在可接受的效能表現上，對於存放在雲端的加密資料提供具使用者隱私保護的詢問能力。
並列摘要
Cloud services become popular in recent years and change how people use the Internet deeply. Many software vendors also utilize the cloud resource to supply greater services for their users. In most of the cloud services, the mechanism of data query always plays an important role. However, during the query process, it usually involves the inspection of what user concerns or even the private data, which should not be disclosed to an untrusted entity, such as a cloud service provider. Maintaining end-user privacy and allowing the query mechanism to work on the cloud seem contradictory. From users’ point of view, there are two kinds of query scenarios. One is the service provider with the readable dataset, like the search engine to its crawled data. The other one is the service provider with the user data which is encrypted. To protect user privacy, the query and the result should be hidden in the former scenario. On the other hand, as the user data is unknown for the service provider in the latter scenario, the user needs to offer a secure index structure. For those scenarios above, we firstly present a framework for building privacy-preserving cloud-based security services. For resource-constrained end-user devices, the security vendors not only provide the fundamental functions for security analysis, but also integrate with cloud-based security services which contain constantly improved security intelligence to assist in protecting against security attacks. With abundant hardware and strong support by security professionals, cloud-based security services can provide better protection than traditional solutions. The framework consists of an architecture for building cloud-based security services and a technique, called private signature filtering, to preserve end-user privacy. The framework supports security monitoring signatures whose correspondence with end-user device queries can be established through conjunction of keywords and numeric value ranges. The framework also allows a trusted middle layer to do a part of the security monitoring computation for the end-user device to reduce the computation overhead on the end-user device. We implement the prototype systems for the cloud-based network intrusion service and the cloud-based malicious URL detection service, to verify effectiveness of our design. The experimental results show that the framework can indeed ensure end-user privacy with acceptable performance overhead in a practical cloud-based security service setting. On the other hand, with the growth in the popularity of cloud storage service (CSS), users can simply store their data in the cloud provided by the CSS and access the data through any device via the Internet anytime and anywhere. For security concern, the accumulation of private data on the cloud requires the use of data encryption to prevent leakage of sensitive information to untrusted third parties. However, as the amount of data kept on the cloud storage is increasing, the use of data encryption makes it difficult or even impossible to locate the data of interest efficiently and securely. Therefore, for the second scenario, we also present a framework for CSS to support queries in encrypted form so that the data on cloud storage can be located efficiently and securely. At the core of the framework is a novel indexing structure, called the bloom filter encrypted search tree (BFEST). The BFEST supports queries in the form of phrase keywords. Client-side encryption, using secret keys that are unknown to the cloud service provider, protects the queries and the retrieved data. We implemented a prototype by extending the hicloud S3 CSS with the proposed framework. The experimental results indicate that the framework can ensure query privacy for encrypted data with an acceptable performance overhead in a practical setting.