Visible to the public Searchable Symmetric Encryption: Designs and Challenges

TitleSearchable Symmetric Encryption: Designs and Challenges
Publication TypeJournal Article
Year of Publication2017
AuthorsPoh, Geong Sen, Chin, Ji-Jian, Yau, Wei-Chuen, Choo, Kim-Kwang Raymond, Mohamad, Moesfa Soeheila
JournalACM Comput. Surv.
KeywordsCloud Security, composability, computing in the encrypted domain, privacy-preserving search, pubcrawl, Resiliency, Searchable encryption
AbstractSearchable Symmetric Encryption (SSE) when deployed in the cloud allows one to query encrypted data without the risk of data leakage. Despite the widespread interest, existing surveys do not examine in detail how SSE's underlying structures are designed and how these result in the many properties of a SSE scheme. This is the gap we seek to address, as well as presenting recent state-of-the-art advances on SSE. Specifically, we present a general framework and believe the discussions may lead to insights for potential new designs. We draw a few observations. First, most schemes use index table, where optimal index size and sublinear search can be achieved using an inverted index. Straightforward updating can only be achieved using direct index, but search time would be linear. A recent trend is the combinations of index table, and tree, deployed for efficient updating and storage. Secondly, mechanisms from related fields such as Oblivious RAM (ORAM) have been integrated to reduce leakages. However, using these mechanisms to minimise leakages in schemes with richer functionalities (e.g., ranked, range) is relatively unexplored. Thirdly, a new approach (e.g., multiple servers) is required to mitigate new and emerging attacks on leakage. Lastly, we observe that a proposed index may not be practically efficient when implemented, where I/O access must be taken into consideration.
Citation Keypoh_searchable_2017