Abstract
dc:descriptionThe demand for cloud storage services continuously increases putting at risk the privacy of the outsourced data. Data encryption is the obvious solution, a well-studied and mature technology. However, it does not support search queries on encrypted data and the server cannot process the outsourced data in their protected form. To counteract this problem, several cryptographic solutions have been proposed, like (Fully) Homomorphic Encryption (FHE), Multi-Party Computation (MPC), Property Preserving Encryption (PPE) and Functional encryption (FE). However, all these technologies are either impractical or weak. One of the most promising cryptographic techniques for searching on encrypted data is Searchable Symmetric Encryption (SSE). SSE schemes offer a nice trade-off between security and efficiency. The primary contribution of the present Phd thesis concentrates on Range and Conjunction Queries using SSE schemes on encrypted data in Key-value databases. We assume that the server is honest-but-curious. Since 2000 and the introduction of SSE from Song et al. [1], most of the research has been concentrated on schemes that support single keyword queries. In this thesis, we investigate the much less explored area of Range Queries SSE. More precisely, we present a way to leverage inverse index computed for single keyword to compute range queries. This is a nice theoretical result that has its own interest. We introduce two SSE schemes that support both single keyword and range queries, building on the new inverted index. First, we present a SSE scheme, called REX, that is no interactive (single round) and no response-hiding scheme. It has optimal communication and search computation complexity, while it is much more secure than traditional Order Preserving Encryption. Next, we present a SSE scheme, called oRQSSE, that supports both single keyword and range queries based on the previous scheme inverted index but it also offers Oblivious Memory Accesses (ORAM), enhancing the security. We prove that our scheme does not leak the search pattern, access pattern, or overlapping pattern except volume pattern. We evaluate its performance, both in theory and in practice. Furthermore, we investigate the area of Conjunctive Queries SSE. We present a SSE scheme that supports conjunctive queries, called ConQSSE. We store popular combinations of keywords consecutive on an inverted index. If the conjunctive query consists of popular combination of keywords then the search and communication complexity are optimal, while if the combination of keywords is not popular then it simplifies each group of consecutive keywords to a single keyword query and send their results to an other scheme that supports conjunctive queries so as to compute the final conjunction. Our scheme is flexible, so it could be combined easily with other scheme and also, it could be implemented and managed using an oblivious memory access (ORAM) algorithm, enhancing it‘s security. The application of our scheme, due to the use of popular combination of keywords, ranges from secure data sharing in healthcare to confidential document retrieval in financial institutions.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Molla, Eirini
- Μολλά, Ειρήνη
Subjects
dc:subject × 13- Κρυπτογραφημένα δεδομένα
- Ασφαλή αναζήτηση
- Αναζήτηση σε κρυπτογραφημένα δεδομένα
- Searchable symmetric encryption
- Encrypted data
- Range queries SSE
- Conjunctive queries SSE
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Υπολογιστές, Υλικό (hardware) και Αρχιτεκτονική
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Computer science, Hardware and Architecture
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/57510