Column-level encryption protects sensitive data at rest but destroys the semantic properties that SQL operators depend on: range comparisons, pattern matching, ordering, and aggregation cease to function over ciphertext. We present a transparent SQL proxy that restores these capabilities using two complementary blind indexes: order hints—keyless bucket-based mappings that preserve approximate order—and n-gram HMACs—keyed truncated MAC sets over character n-grams that enable substring containment checks, with optional per-n-gram random nonces that eliminate cross-row correlation. A two-phase query pipeline rewrites SQL with widened operators over auxiliary columns (Phase 1, database-side, zero false negatives by construction), then decrypts and verifies results against original predicates (Phase 2, proxy-side). We formally prove that order hint leakage is strictly less than Order-Preserving Encryption (OPE), placing our scheme in a precisely defined leakage hierarchy. We evaluate the system across four SQL dialects (PostgreSQL, MySQL, SQL Server, Oracle) on a ∼ 1M-row medical records workload, demonstrating sub-millisecond latency for selective queries and under 2 × overhead for multi-table joins, with pattern-matching queries performing comparably to cleartext.
Maintaining access control policies is an ongoing process to ensure required but not excessive authorizations. Organizations thus leverage various data sources to ease this maintenance. Among these data sources are access control matrices, attributes, access logs, and transaction logs. While research reasonably covers the former data sources, the potential of transaction logs remains untapped. We pave the way for transaction logs as a data source in access control by (i) expressing them with a formalization, (ii) pinpointing them in typical Identity and Access Management (IAM) infrastructures, and (iii) grounding them in IAM processes. We conclude that access control transaction logs are valuable data sources for improving analytical capabilities for IAM.
The rapid growth of decentralized systems in the Web3 ecosystem has introduced numerous challenges, particularly in ensuring data security, privacy, and scalability [3, 8]. These systems rely heavily on distributed architectures, requiring robust mechanisms to manage data and interactions among participants securely. One critical aspect of decentralized systems is key management, which is essential for encrypting files, securing database segments, and enabling private transactions. However, securely managing cryptographic keys in a distributed environment poses significant risks, especially when nodes in the network can be compromised [9]. This research proposes a decentralized database scheme specifically designed for secure and private key management. Our approach ensures that cryptographic keys are not stored explicitly at any location, preventing their discovery even if an attacker gains control of multiple nodes. Instead of traditional storage, keys are encoded and distributed using the BFLUT (Bloom Filter for Private Look-Up Tables) algorithm [7], which enables secure retrieval without direct exposure. The system leverages OrbitDB [4], IPFS [1], and IPNS [10] for decentralized data management, providing robust support for consistency, scalability, and simultaneous updates. By combining these technologies, our scheme enhances both security and privacy while maintaining high performance and reliability. Our findings demonstrate the system's capability to securely manage keys, prevent unauthorized access, and ensure privacy, making it a foundational solution for Web3 applications requiring decentralized security.
The notion that federated learning ensures privacy simply by keeping data local is widely acknowledged to be flawed. Cryptographic techniques such as Multi-Party Computation (MPC) and Fully Homomorphic Encryption (FHE) address this issue by concealing the model during the training procedure, but their extreme computational and communication overhead makes them impractical for real-world deployment. However, we argue that such strong guarantees are unnecessary. Even with full-model encryption, black-box attacks remain possible during the prediction phase, since model outputs are eventually revealed to the querier. This suggests that instead of enforcing perfect privacy during training, it is sufficient to ensure that the leakage during training is no higher than the leakage during prediction. To achieve this, we generalize POSEIDON (NDSS 2021), a state-of-the-art FHE-based federated learning approach, by selectively encrypting only the components of the model necessary to match the privacy level of the prediction phase. Our method identifies the parts of the model that contribute most to information leakage and prioritizes their encryption, significantly reducing computational and communication overhead. Our experiments on dense neural networks show that encrypting only the last layer is often sufficient to hinder white-box attacks, improving efficiency by a linear factor in the number of layers. For deeper models, multiple layers may require encryption, but our approach still achieves a substantial speedup compared to full-model encryption.
Any organization providing services to users typically requires the users to share some degree of personal information in order to access these services. However, users seldom have complete visibility or control over their own data, justifiably raising serious privacy concerns. Additionally, organizations are often inhibited from collaboration due to the increased risk of privacy breaches, and corresponding reporting requirements. Increasing privacy focused legislation worldwide such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) create complex protection and reporting requirements that need to be systematically enforced. The Right To Be Forgotten (RTBF) recommendations included in GDPR essentially demand that every individual has the right to request for erasure of data pertaining to it, and thus in effect be forgotten from the system on demand. We propose a novel approach called CRISP (Consensus-enabled, Redactable, Immutable, Securely shareable, Provable) that uses permissioned enterprise blockchains for trustless interoperation among a network of identifiable organizations providing the ability to support RTBF. It combines off-chain data storage and distributed resource sharing with blockchain consensus mechanisms and decentralized access control. We describe the various components of CRISP and delineate the detailed steps of its operation. Results of an extensive set of experiments clearly establish the viability of our approach.
