d-DSE: Distinct Dynamic Searchable Encryption Resisting Volume Leakage in Encrypted Databases

Core Concepts

Developing a new perspective on handling volume leakage in encrypted databases through distinct search with the d-DSE scheme.

Abstract

新しい視点でボリューム漏洩を処理するためのDistinct Dynamic Searchable Encryption（d-DSE）スキームを開発しました。このスキームは、Bloom Filter（BF）とSymmetric Revocable Encryption（SRE）を組み合わせて大規模なデータセットに対処し、非インタラクティブな削除とサブリニア検索を実現しています。BF-SREはFP＆BPおよびDwVHセキュリティを達成し、効果的なデータ保護を提供します。

Stats

MONETA [5]の計算コストはO(aw logN +log3 N)であり、通信コストはO(log2 N)です。
DIANAdel [5]の計算コストはO(aw)であり、通信コストはO(logaw)です。
JANUS [5]の計算コストはO(nw ·dw)であり、通信コストはO(1)です。
AURA [53]の計算コストはO(nw)であり、通信コストはO(1)です。
MITRA [21]の計算コストはO(aw)であり、通信コストはO(1)です。
SEAL* [16]では静的データベースに焦点が当てられており、更新操作をサポートしていませんが、キーワードによる値の取得が可能です。また、クライアントの状態をローカルに保存します。

Quotes

"Padding strategies are popularly applied to mitigate the impact of volume leakage in EDBs."
"A new perspective - starting from distinct search."
"We propose the Distinct Dynamic Searchable Symmetric Encryption (d-DSE) that enables clients to securely search for distinct values with volume-hiding."

Key Insights Distilled From

d-DSE

by Dongli Liu,W... at arxiv.org 03-05-2024

https://arxiv.org/pdf/2403.01182.pdf

Deeper Inquiries

How does BF-SRE compare to other schemes in terms of security and efficiency

BF-SRE stands out in terms of security and efficiency compared to other schemes. In terms of security, BF-SRE offers Forward Privacy (FP) and Backward Privacy (BP), ensuring that the update and search functions are protected from privacy breaches. Additionally, BF-SRE provides Distinct with Volume-Hiding (DwVH) security, preventing adversaries from distinguishing signatures in the encrypted database. This robust security framework makes BF-SRE a reliable solution for protecting data in encrypted databases.
Efficiency-wise, BF-SRE utilizes Bloom Filters (BF) to minimize storage costs while efficiently managing large-scale datasets. The integration of Symmetric Revocable Encryption (SRE) enables non-interactive deletion, enhancing operational efficiency in handling updates and deletions within the database. Overall, BF-SRE strikes a balance between strong security measures and efficient data management capabilities.

What are the potential limitations or vulnerabilities of the d-KW-dDSE construction

One potential limitation or vulnerability of the d-KW-dDSE construction lies in its susceptibility to pattern leakage attacks. Since d-KW-dDSE involves storing distinct values using hash maps on the client side, there is a risk of revealing patterns related to keyword searches and value quantities during query executions. Adversaries could potentially exploit these patterns to infer information about the dataset structure or perform targeted attacks based on leaked patterns.
Additionally, d-KW-dDSE may face challenges related to scalability when dealing with extremely large datasets or complex relational structures. Managing hash maps for each keyword-value pair can become resource-intensive as the dataset grows, potentially impacting performance and storage requirements.
To mitigate these limitations, careful consideration should be given to optimizing data structures and encryption mechanisms used in d-KW-dDSE to enhance resilience against pattern leakage attacks and ensure scalability across diverse database environments.

How can the concept of DwVH security be applied to other encryption schemes beyond d-DSE

The concept of Distinct with Volume-Hiding (DwVH) security can be applied beyond d-DSE to enhance confidentiality and privacy protection in various encryption schemes. By incorporating DwVH principles into other encryption frameworks such as Searchable Symmetric Encryption (SSE), Attribute-Based Encryption (ABE), or Homomorphic Encryption, it is possible to prevent adversaries from inferring volumetric information through pattern analysis or statistical inference techniques.
For example:

In SSE: DwVH can help conceal volume-related patterns during keyword searches within encrypted databases.
In ABE: DwVH can safeguard attribute retrieval operations by hiding volume correlations among attributes.
In Homomorphic Encryption: DwVH can protect computation results by concealing volume-related characteristics that could reveal sensitive information about encrypted data sets.
By integrating DwVH principles into diverse encryption schemes, organizations can strengthen their data protection strategies against volumetric attacks while maintaining high levels of confidentiality and privacy across different cryptographic applications.