Searchable Encrypted Relational Databases: Risks and Countermeasures

Abdelraheem, Mohamed Ahmed; Andersson, Tobias; Gehrmann, Christian

doi:10.1007/978-3-319-67816-0_5

Cited by 5 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4) Count and Relational-count Attack: The count attack is proposed by Cash et al in [12] to recover encrypted queries in file storage systems based on the access and size patterns leakage. In [36], Abdelraheem et al have applied this attack to databases and named it a relational-count attack. As in the IKK attack scenario, the CSP is also assumed to know an m × m matrix C, where its entry C[w i , w j ] holds the cooccurrence rate of keyword w i and w j in the targeted dataset.…”

Section: B Attacks Against Se Solutionsmentioning

confidence: 99%

Privacy-preserving Dynamic Symmetric Searchable Encryption with Controllable Leakage

Cui

Song

Asghar

et al. 2021

ACM Trans. Priv. Secur.

View full text Add to dashboard Cite

Searchable Encryption (SE) is a technique that allows Cloud Service Providers to search over encrypted datasets without learning the content of queries and records. In recent years, many SE schemes have been proposed to protect outsourced data. However, most of them leak sensitive information, from which attackers could still infer the content of queries and records by mounting leakage-based inference attacks, such as the count attack and file-injection attack . In this work, first we define the leakage in searchable encrypted databases and analyse how the leakage is leveraged in existing leakage-based attacks. Second, we propose a <underline>P</underline>rivacy-preserving <underline>M</underline>ulti-<underline>c</underline>loud based dynamic symmetric SE scheme for relational <underline>D</underline>ata<underline>b</underline>ase ( P-McDb ). P-McDb has minimal leakage, which not only ensures confidentiality of queries and records but also protects the search, intersection, and size patterns. Moreover, P-McDb ensures both forward and backward privacy of the database. Thus, P-McDb could resist existing leakage-based attacks, e.g., active file/record-injection attacks. We give security definition and analysis to show how P-McDb hides the aforementioned patterns. Finally, we implemented a prototype of P-McDb and tested it using the TPC-H benchmark dataset. Our evaluation results show that users can get the required records in 2.16 s when searching over 4.1 million records.

show abstract

Section: B Attacks Against Se Solutionsmentioning

confidence: 99%

Privacy-preserving Dynamic Symmetric Searchable Encryption with Controllable Leakage

Cui

Song

Asghar

et al. 2021

ACM Trans. Priv. Secur.

View full text Add to dashboard Cite

show abstract

“…Specifically, the searchable encryption approach is an improvement from Mi-moSecco [9] and the vertical fragmentation is done via a pre-defined set of privacy constraints in order to provide protection against inference attacks on searchable encryption schemes. The DB proxy for secure data outsourcing, trading-off performance and security, is presented in [10], and in [11] vertical fragmentation is added as a privacy-related countermeasure to create the current PaaSword scheme. Moreover, a simple and efficient searchable symmetric encryption (SSE) scheme capable of executing Boolean queries (i.e., multi-keyword queries) on an encrypted Bitmap index is proposed in [12].…”

Section: Paasword's Scientific Contributionmentioning

confidence: 99%

“…A PEP represents either a policy or a policy set containing rules and an entry point of validation and enforcement. Within a Spring project 10 , this annotation can also be used together with an interception mechanism based on Spring AOP 11 . PaaSwordDDE is a class-level annotation for Data Distribution and Encryption.…”

Section: Third-party Layermentioning

confidence: 99%

“…However, in PaaSword, we propose using them to strengthen the security of the SSE schemes against inference attacks. More details about the security gain when distributing a searchable encrypted relational database using vertical fragmentation via pre-defined privacy constraints, can be found in [11].…”

Section: Deterministic Index Encryptionmentioning

confidence: 99%

“…-Forward privacy schemes: Forward privacy means that an update procedure should not leak if a newly inserted record or document matches previous search queries [49]. Such a property prevents the recent devastating adaptive file or record injection attacks on SSE schemes [11,50]. Another enabled feature about forward private schemes is that they allow the secure and on-line construction of an encrypted database index (this is to be contrasted with the PaaSword approach whereby the encrypted database index is constructed off-line and then outsourced to the cloud).…”

Section: Data Privacymentioning

confidence: 99%

See 2 more Smart Citations

PaaSword: A Data Privacy and Context-aware Security Framework for Developing Secure Cloud Applications - Technical and Scientific Contributions

Papatheocharous

Mantzouratos

Gouvas

et al. 2017

European Space Projects: Developments, Implementations and Impacts in a Changing World

Self Cite

View full text Add to dashboard Cite

Most industries worldwide have entered a period of reaping the benefits and opportunities cloud offers. At the same time, many efforts are made to address engineering challenges for the secure development of cloud systems and software. With the majority of software engineering projects today relying on the cloud, the task to structure end-to-end secure-by-design cloud systems becomes challenging but at the same time mandatory. The PaaSword project has been commissioned to address security and data privacy in a holistic way by proposing a context-aware security-by-design framework to support software developers in constructing secure applications for the cloud. This chapter presents an overview of the PaaSword project results, including the scientific achievements as well as the description of the technical solution. The benefits offered by the framework are validated through two pilot implementations and conclusions are drawn based on the future research challenges which are discussed in a research agenda.

show abstract