homomorphic encryption

This work addresses the challenge of verifying the integrity of results returned by cloud servers in dynamic encrypted control systems by proposing a lightweight verification mechanism. It introduces linear dynamical system theory into encrypted control verification for the first time, enabling efficient local validation of computational correctness. The approach operates by injecting artificial challenge signals in parallel at the cloud side and leveraging the inherent input-output system properties of the controller. The proposed scheme incurs negligible additional computational overhead, detects tampering errors with high probability, and effectively mitigates replay attacks. Consequently, it significantly outperforms existing static or computationally expensive verification methods.

This work addresses the challenge of performing range counting over multi-party distributed geospatial data while simultaneously preserving query privacy, ensuring computational efficiency, and maintaining accuracy in the presence of overlapping data—goals that existing methods struggle to reconcile. To this end, the paper proposes the PPRC protocol, which for the first time achieves a unified optimization of these three objectives. PPRC leverages two key techniques: Private Range Predicates (PRP) and Oblivious Linear Counting (OLC), replacing secure comparison with encrypted membership testing and employing lightweight cryptographic operations to enable efficient and secure range evaluation and aggregation. Experimental results on both real-world and synthetic datasets demonstrate that PPRC reduces estimation error by up to 55× and improves runtime performance by as much as 37× compared to baseline approaches.

To address privacy risks arising from access pattern leakage in encrypted range queries under cloud environments, this paper proposes a novel scheme that jointly achieves strong security guarantees and high efficiency. Methodologically, it introduces, for the first time, a learnable spatial index into encrypted settings, integrating Paillier homomorphic encryption with a hierarchical prediction architecture. It further designs a noise-injected bucket mechanism and a permutation-based secure bucket prediction protocol, augmented by a secure point extraction protocol, to simultaneously protect data confidentiality, query content, and access patterns. Experimental evaluation on both real-world and synthetic datasets demonstrates that the proposed scheme significantly outperforms state-of-the-art approaches in query latency and throughput, while providing rigorous formal security proofs under standard cryptographic assumptions.

Existing searchable encryption schemes struggle to simultaneously support Boolean range queries over spatial data while preserving both access and search pattern privacy. This work proposes BRASP, the first scheme to achieve strong privacy guarantees for Boolean range queries on encrypted spatial data. BRASP constructs an encrypted inverted index using Hilbert curve prefix encoding and, under a dual non-colluding server architecture, integrates index shuffling, ID field redistribution, and a forward-secure mechanism to effectively conceal query patterns while supporting dynamic updates. Experimental evaluation on real-world datasets demonstrates that BRASP achieves low computational and communication overhead alongside high practicality. To ensure reproducibility, the implementation has been made publicly available.

This work addresses a critical limitation in existing certification schemes for encrypted machine learning models, which only verify model behavior on a fixed audit dataset and thus fail to guarantee generalization to new, identically distributed data—rendering them vulnerable to adversarial manipulation. We formally introduce, for the first time, a generalizable security definition tailored to encrypted model certification and expose fundamental assumptions underlying current zero-knowledge proof–based privacy-preserving auditing protocols that do not hold in practical deployments. To bridge this gap, we propose a unified certification framework integrating secure multi-party computation, zero-knowledge proofs, and statistical generalization theory, providing formal guarantees that audit outcomes generalize to real-world data. Empirical evaluation demonstrates that adversaries can achieve over 99% accuracy during audits while degrading true model performance to below 30%; our protocol effectively mitigates such attacks, aligning theoretical assurances with real-world robustness.

Dynamic encrypted control faces a common challenge: degradation of closed-loop stability due to homomorphic encryption noise accumulation and arithmetic overflow in encoded representations. This paper presents the first unified stability analysis and performance evaluation of four mainstream approaches—bootstrapping, periodic state reset, integer-domain reconstruction, and FIR-based controllers—under identical benchmark system conditions. Leveraging both numerical simulation and Lyapunov stability theory, we quantitatively characterize the trade-offs among control accuracy, closed-loop stability, and computational overhead. Results show that bootstrapping effectively suppresses noise but incurs prohibitive computational cost; periodic state reset achieves a favorable balance between stability and efficiency; integer-domain reconstruction improves precision yet suffers from limited dynamic range; and FIR controllers inherently avoid feedback-induced noise accumulation. Our analysis provides both theoretical foundations and practical guidelines for designing and selecting privacy-preserving control architectures.

This work addresses a critical gap in existing encrypted geo-search and zero-knowledge proximity proof systems, which fail to link proximity proofs to their original searches after session state erasure, thereby undermining authorized provenance. We propose Search-Bound Proximity Proofs (SBPP), which securely bind search authorization to proofs by decoupling session nonces, Merkle-root commitments over result sets, and signed receipts into independently auditable components—without modifying the underlying ZKP circuitry. We formalize, for the first time, a security model for Search Authorization Proofs (SAP), identifying and mitigating forensic misattribution attacks arising from cross-session parameter reuse, and enabling attribute-level fault isolation in offline audits. Evaluated on 110,776 OpenStreetMap points of interest across real-world and synthetic datasets, our approach incurs only sub-millisecond overhead over a 125 ms baseline.

Democracies are built upon secure and reliable voting systems. Electronic voting systems seek to replace ballot papers and boxes with computer hardware and software. Proposed electronic election schemes have been subjected to scrutiny, with researchers spotting inherent faults and weaknesses. Inspired by physical voting systems, we argue that any electronic voting system needs two essential properties: ballot secrecy and verifiability. These properties seemingly work against each other. An election scheme that is a complete black box offers ballot secrecy, but verification of the outcome is impossible. This challenge can be tackled using standard tools from modern cryptography, reaching a balance that delivers both properties. This tutorial makes these ideas accessible to readers outside electronic voting. We introduce fundamental concepts such as asymmetric and homomorphic encryption, which we use to describe a general electronic election scheme while keeping mathematical formalism minimal. We outline game-based cryptography, a standard approach in modern cryptography, and introduce notation for formulating elections as games. We then give precise definitions of ballot secrecy and verifiability in the framework of game-based cryptography. A principal aim is introducing modern research approaches to electronic voting.

This work addresses the challenge of uniformly modeling information leakage arising from persistent hidden states in interactive cryptographic systems by proposing a security framework grounded in observer quotient spaces. It introduces session-tagged observer-indexed experiments, adaptive schedulers, and ideal quotient functionalities to explicitly couple hidden-state persistence with observer capabilities, thereby establishing composable leakage bounds and reducing security to indistinguishability between real and ideal worlds. The paper innovatively formulates additive sequential flaws, product total-variation parallel bounds, and an adaptive observer encapsulation mechanism. Leveraging non-uniform simulators, observability kernel identification, and control-theoretic tools—such as dissipativity and input-to-state stability (ISS) residual bounds—it achieves advantage reduction. The framework’s efficacy is validated through leakage auditing and linear time-invariant (LTI) observer design in settings including IND-CPA encryption, deterministic encryption, and timing, cache, power, and electromagnetic side channels.

This work addresses the limitations of traditional runtime monitoring systems, which rely on a single monitor and heavyweight cryptographic techniques, thereby struggling to balance privacy preservation with real-time scalability. The paper proposes the first secret-sharing protocol supporting continuous monitoring, replacing conventional encryption with a distributed multi-party architecture that assumes at least one honest participant. This approach ensures strong privacy guarantees while maintaining a dynamic internal state. Implemented within the MP-SPDZ framework, the system demonstrates significantly reduced computational overhead and outperforms existing solutions in both scalability and performance, making it well-suited for real-time monitoring applications.