文件複製及分割為基礎之隱私保護多關鍵字搜尋

Efficient Privacy-Preserving Multi-keyword Ranked Search Utilizing Document Replication and Partition

曾清陽(Ching-Yang Tseng)

指導教授 ： 周承復

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.10381

摘要

近年來，雲端服務已經逐漸走進我們日常生活中，越來越多資料開始集中到公開的儲存器上。然而，在公開的環境下私密資料可能會遭受來自外部或內部的威脅，故在上傳前必須經過加密。但透過傳統的加密方式會造成無法提供如明文搜尋上的功能，而在過去的方法裡大多只能提供單關鍵字或布林關鍵字搜尋，嚴重地的限制了使用者的搜尋功能且搜尋結果不佳。最近，Cao等人發表了MRSE，是第一篇在加密環境下提供多關鍵字搜尋且依關聯性排序結果。而由於MRSE加密方法以及文章排名的方式，當需更新新的文件時需要重建整個索引才能達到目標，這對在現實生活中，雲端儲存器經常需要更新新的資料會造成很大的負擔。
在本篇論文中，我們利用文件複製及分割來保護搜尋者的行為避免雲端伺服器的窺探，同時利用我們提出的索引組結構來讓改善更新效率不佳的問題。而在有威脅的環境中的系統運行，使用者用來搜尋的訊息及雲端儲存器上的索引和資料區塊皆不會洩漏與資料有關的額外訊息(如:關鍵字)，雖然在使用者端需要多負擔一點點的計算量，但仍能在高度安全性保證下進行top-k多關鍵字搜尋。

關鍵字

隱私保護 ； 關鍵字搜尋 ； 排序搜尋 ； 文件複製 ； 文件分割

並列摘要

Nowadays, the widespread notion of cloud computing is penetrating into our daily life. However, the data on public cloud storage may incur privacy disclosure since the placement is not under users' trusted domain. Therefore, sensitive data should be encrypted before uploaded to cloud servers. Unfortunately, traditional symmetric encryption on data will restrict the functionality as in plaintext keyword search. Most of previous works on searchable encryption provide single keyword search or Boolean keyword search which either restricting the users' operation on data or retrieving out-of-order relevance documents.
Recently, Cao et al. propose MRSE cite{Cao} which is the first work that addresses the problem of multi-keyword search with ranked result on encrypted cloud data. Nevertheless, the predefined keyword dictionary will greatly reduce the flexibility of updating the system. In this work, for the first time we utilize the document replication and partition to protect users' behaviour from the monitor of cloud server. Then, we propose a new index structure so that the data owner can update his collection efficiently. Also, with the minor overhead on user side, users are still able to perform multi-keyword search efficiently with top-k relevant result list and provide strong security guarantee at the same time. Finally, our evaluation results show that the proposed system operates competitive query performance and introduce lower update overhead with the state of the art work.

並列關鍵字

privacy-preserving ； keyword search ； ranked search ； document replication ； document partition

參考文獻

key encryption with keyword search. In Christian Cachin and Jan Camenisch,

multi-keyword ranked search over encrypted cloud data. In INFOCOM 2011. 30th

IEEE International Conference on Computer Communications, Joint Conference of

China, pages 829--837. IEEE, 2011.

[2] Chien-Kang Huang, Lee-Feng Chien, and Yen-Jen Oyang. Relevant term suggestion