Hasina Andriambelo, Hery Zo Andriamanohisoa, Naghmeh Moradpoor
Federated learning enables collaborative model training without sharing raw data, but practical deployments increasingly require verifiable guarantees that clients compute updates correctly. Zero-knowledge proofs can provide such guarantees, yet existing approaches face scalability limits due to the combined cost of polynomial commitments and fast Fourier transform (FFT) intensive verification. Pairing-based schemes offer compact proofs but incur high prover and verifier overhead, while hash-based constructions reduce algebraic cost at the expense of rapidly growing proof sizes. This paper proposes Hybrid-Commit, a polynomial commitment architecture for Binius zero-knowledge proofs that aligns cryptographic primitives with the algebraic structure of federated learning workloads. The scheme separates verification into additive and multiplicative phases: linear aggregation is handled using batched additive commitments optimized for binary fields, while non-linear constraints are verified via hash-based commitments over sparsely selected FFT domains. Proofs from multiple clients are combined through recursive aggregation while preserving non-interactivity. Experiments demonstrate scalability in prover time and proof size (near-constant prover time across 4–11 clients; 160 bytes per client representing 341× and 813× reductions vs. FRI-PCS and Orion), although verification time (762 ms per client) does not scale favorably, making the scheme suitable for bandwidth-constrained scenarios. The scheme achieves under 2% end-to-end training overhead with no impact on model accuracy, indicating that workload-aware commitment design can improve specific scalability dimensions of zero-knowledge verification in federated learning systems.
The Internet of Medical Things (IoMT) enables real-time health monitoring and intelligent clinical decision-making by continuously collecting and processing sensitive physiological data from wearable, implantable, and edge-connected devices. However, this data aggregation paradigm introduces critical privacy and security challenges, including data leakage, aggregator misbehavior, and adversarial attacks, while existing frameworks often fail to simultaneously ensure confidentiality, verifiability, and efficiency. To address these limitations, we propose MedGuard, a novel end-to-end secure data aggregation framework for IoMT that synergistically integrates Fully Homomorphic Encryption (FHE) based on the CKKS scheme and Groth16 zero-knowledge Succinct Non-Interactive Arguments of Knowledge (zk-SNARKs). MedGuard enables healthcare providers to perform complex analytical queries, such as statistical analysis, anomaly detection, and trend forecasting, directly on encrypted data without decryption, ensuring compliance with privacy regulations. By allowing edge nodes to generate cryptographic proofs of correct computation and enabling cloud-based verification, MedGuard eliminates reliance on trusted intermediaries and mitigates insider threats. Our comprehensive evaluation, conducted in a high-fidelity OMNeT++ 6.0.1 simulation environment with 1,000 IoMT devices, 100 edge nodes, and an Amazon EC2 c5.4xlarge cloud server, uses a hybrid dataset combining real-world and GMM-augmented synthetic data. Results show that MedGuard achieves an end-to-end latency of 64.8 ms, a 13.3% improvement over state-of-the-art baselines, communication efficiency of 1.465 GB/s, per-query energy consumption of 1.489 mJ, and sustained throughputs of 1,200 packets/s, 120 aggregates/s, and 1,200 queries/s. These performance gains, combined with a robust [Formula: see text] security level, demonstrate that MedGuard delivers scalable, verifiable, and privacy-preserving analytics for next-generation smart healthcare systems.
We present Y.I.N.-LLM, a privacy-preserving training architecture for Large Language Models that mathematically guarantees non-memorization of training data. The core innovation is the mandatory DP→ZK→HE ordering (Differential Privacy → Zero-Knowledge Proof → Homomorphic Encryption) applied to transformer gradients during training. Key results: (1) 2.3% accuracy loss at ε=1.0 privacy versus 15-40% with standard DP-SGD; (2) zero extractable training data across all tested attack vectors; (3) native GDPR Article 17 "right to be forgotten" compliance via cryptographic gradient subtraction; (4) EU AI Act Article 50 transparency compliance through verifiable privacy proofs. The Non-Memorization Theorem establishes that for any model M trained with Y.I.N.-LLM parameters (ε, δ), the probability of verbatim reproduction is bounded: P[M outputs y | x ∈ training] ≤ e^ε · P[M outputs y | x ∉ training]. This transforms copyright defense from argument to mathematics. Y.I.N.-LLM addresses the $10B+ memorization litigation crisis (NYT v. OpenAI, Getty v. Stability AI, Authors Guild v. OpenAI) by providing the first mathematically verifiable non-memorization guarantee with practical accuracy preservation. Patent Protected: U.S. Provisional Application 63/946,118 (filed December 21, 2025).
Federated Learning enables large-scale collaborative training across distributed devices. However, in massive-scale Internet-of-Things (IoT) deployments, ensuring the trustworthy sensor-level operations remains a critical challenge. We introduce a hierarchical framework that combines a three-tier architecture (devices → gateways → server) with a high-speed recursive proof system to enforce scalable zero-knowledge proofs (ZKPs). At the device level, each proof serves as a unified cryptographic commitment, binding the device’s identity, local data integrity, and training correctness into a single attestation. These proofs are then individually verified at intermediate gateways, and compressed into a single, succinct proof using a folding scheme inspired by Nova [1] - a state-of-the-art system that can excel at this task at best. The server then verifies a small number of batched proofs before aggregation, reducing workload (∼ 571× in data load) by replacing hundreds of thousands of individual proof and model update transmissions with just one per gateway. Our fully implemented R1CS precursor demonstrates resilience against various vectors (e.g., backdoor-style attacks,) achieves a ∼ 34× verification speedup on a 105-device network, and maintains both strong security and model performance. Our prototype, evaluated on an Internet-of-Vehicles (IoV) use case, demonstrates that recursive proofs add succinct overhead while providing scalable, robust integrity guarantees against adversarial environments.
Federated learning enables multiple data owners to collaboratively train a global model, but verifying the correctness of submitted updates remains a critical challenge. Malicious clients may poison model weights, and colluding validators can undermine aggregation. We propose DI-FLAME, a decentralized validation framework that introduces a stake-and-proof mechanism for model verification. Each claim is submitted with a justification and stake, evaluated via a black-box credibility function. Validators engage in peer review, and contradictors may challenge weak claims by submitting stronger proofs and higher stakes. Rewards and penalties are dynamically distributed based on justification strength and challenge outcomes. DIFLAME provides robust defense against poisoned updates, even under adversarial majorities. While blockchain infrastructure is not required, DI-FLAME is compatible with with decentralized ledgers, enabling transparent recording of validation outcomes when deployed over a blockchain.
Dr.B.Lakshma reddy, Chetan Kalamadi, B GURURAJ, Dhruva dinesh naik
This research presents Agri Safe, an innovative framework combining distributed ledger technology with machine intelligence for agricultural property documentation. Agricultural sustainability depends heavily on clear property rights, yet existing documentation mechanisms suffer from vulnerabilities enabling unauthorized modifications. Our solution leverages immutable distributed ledgers to ensure record permanence while employing classification algorithms including support vector machines, logistic regression, and random forests for pre-validation fraud detection. This dual-layer approach prevents illegitimate entries from consuming computational resources. Validation rules embedded within automated contracts ensure operational efficiency and data integrity. The Inter Planetary File System handles supplementary documentation, with cryptographic hashes anchored on-chain for verification. Comprehensive evaluation demonstrates system reliability through AI model performance assessment and smart contract security analysis. Agri Safe establishes trust by eliminating unauthorized modifications and minimizing property conflicts.
Federated Learning (FL) has emerged as a transformative paradigm for distributed machine learning, enabling model training across decentralized edge devices while preserving data privacy. This methodology is critical for sectors handling sensitive information, such as finance, healthcare, and the Internet of Things (IoT). Despite its benefits, the coordination and communication overhead between distributed nodes remain significant challenges. This paper evaluates the efficacy of REST and GraphQL API architectures in facilitating FL workflows. While REST APIs are favored for their statelessness and simplicity, GraphQL offers enhanced flexibility and efficiency by enabling precise data fetching—a vital feature for bandwidth-constrained decentralized systems. We provide a comparative analysis of these paradigms across performance, security, and scalability metrics, specifically regarding data synchronization and model aggregation. Finally, we propose design best practices for developing APIs that support robust, compliant, and efficient federated prediction systems.
Zuodong Wu, Dawei Zhang, Mianxiong Dong, Kaoru Ota
In the data-driven era, the unchecked collection and processing of personal data has given rise to serious privacy concerns. In response, the General Data Protection Regulation (GDPR) was introduced to grant individuals stronger control over the use of their data. Privacy data retrieval methods show considerable promise in this context, but further improvements are required to balance the principles of lawfulness and data minimization. To address this problem, we propose PDRAA, an efficient privacy data retrieval protocol with anonymous authorization based on the verifiable credential (VC). Specifically, our designed VC achieves anonymous identification of data subjects and facilitates fine-grained access control by supporting selective disclosure of attributes. By combining VC with non-interactive zero-knowledge (NIZK) proofs, PDRAA enables data subjects to anonymously authenticate via VC presentation. This allows the data controller to verify the legitimacy of retrieval requests while ensuring compliance with the principle of data minimization. Besides, PDRAA introduces a re-randomization mechanism to prevent linkability attacks during the authorization process and provides lightweight, flexible authorization revocation. Moreover, we utilize Labeled Private Set Intersection (Labeled PSI) technology to meet the privacy requirements of participants and support batch retrieval. Our protocol takes a comprehensive security analysis within the Universal Composability framework. Experimental results demonstrate that PDRAA outperforms existing methods in terms of performance, which is significant for promoting compliance with GDPR.
The convergence of vehicular technology, artificial intelligence (AI), and distributed computing has catalyzed the emergence of the Internet of Vehicles (IoV) as a cornerstone of next-generation intelligent transportation systems (ITS). By enabling vehicle-to-everything (V2X) communication, IoV supports cooperative perception, real-time decision-making, and autonomous driving. However, the reliance on large-scale, data-driven intelligence in IoV exposes systems to critical challenges, including adversarial poisoning, privacy leakage, identity forgery, and the fragility of centralized learning architectures. Federated Learning (FL) has been proposed as a promising paradigm to alleviate some of these issues by enabling distributed model training without centralizing sensitive vehicular data. Nonetheless, conventional FL remains vulnerable to security and trust limitations, particularly in dynamic vehicular environments. This thesis addresses these challenges by designing secure, privacy-preserving, and scalable FL frameworks that leverage distributed ledger technologies and cutting-edge security mechanisms.The thesis advances knowledge through four interconnected contributions. First, two novel optimization-driven poisoning attack models are introduced: PA-PSOSA and PAPSOGA, which combine particle swarm optimization with simulated annealing and genetic algorithms, respectively. These models demonstrate that even a small poisoning budget can substantially degrade global model utility under black-box and clean-label constraints, highlighting the urgency of robust defenses in vehicular FL. Second, a permissioned blockchain-enabled FL (BCFL) framework is proposed, in which consortium edge nodes running Practical Byzantine Fault Tolerance (PBFT) consensus replace the central aggregator. With blockchain integration and data validation mechanisms, this design ensures identity authentication, verifiable audit trails, and improved resilience against poisoning and Sybil attacks, while maintaining high model accuracy under adversarial conditions. Third, the framework is further enhanced to achieve inference-resistance by integrating secure aggregation (SecAgg) and differential privacy (DP), and lightweight with off-chain commitments. This design significantly reduces ledger storage requirements, increases system throughput, and mitigates inference-based privacy risks. Finally, to overcome the scalability limitations of PBFT-based BCFL, a DAG-enabled FL (DFL) framework is developed. By leveraging parallel validation, utility-score-based tip selection, and reputation-weighted aggregation, this framework significantly improves scalability, reduces communication complexity, and enhances robustness in asynchronous vehicular environments.Together, these contributions articulate a coherent progression from exposing vulnerabilities in vehicular FL to constructing secure, privacy-preserving, and scalable frameworks tailored for IoV ecosystems. The findings demonstrate that interdisciplinary integration of optimization theory, cryptography, differential privacy, and distributed ledger technologies is indispensable for trustworthy vehicular intelligence. Beyond theoretical significance, the proposed frameworks offer practical designs for deployment in safety-critical IoV environments. Future research directions include the integration of zero-knowledge proofs (ZKP) for verifiable privacy, adaptive defenses against evolving adversarial strategies, and experimental validation in real-world vehicular testbeds. Collectively, this thesis establishes a foundation for secure federated intelligence in IoV, contributing to the reliability, efficiency, and trustworthiness of next-generation ITS.
A protocol-level transaction tax faces a structural privacy tradeoff: jurisdictional attribution requires releasing aggregates from which an adversary with auxiliary information may attempt to learn properties of the underlying transactions. This paper formalizes the tradeoff and proves two mechanism-level theorems. Theorem 1 establishes that the composition of Pedersen commitments with zero-knowledge validity proofs preserves individual-transaction hiding under standard cryptographic assumptions, with no distributional requirement on transaction amounts. Theorem 2 separates two cohortaggregation threats that prior analyses have conflated and bounds each. Theorem 2a (event detection): against an adversary attempting to detect that an anomalous transaction of magnitude 𝜏 occurred in a cohort of size 𝑘, the optimal likelihood-ratio-test.
The Internet of Vehicles (IoV) cloud platform enables multi-dimensional sharing of vehicle data, enhancing both its value and the quality of vehicle services. To ensure dynamic authorized access control and safeguard user privacy in the IoV cloud platform, an attribute-based encryption (ABE) scheme is employed. However, existing state-of-the-art schemes still struggle to simultaneously address the following challenges: 1) the limitation of single-application scenarios due to the difficulty of achieving cross-primitive ciphertext transformation; 2) the risk of user privacy leakage caused by the inability to fully hide access policies; 3) the potential for malicious accusations against the delegator due to a lack of verifiability. To tackle these issues, we propose PPVTAC, a Privacy-Preserving, Verifiable, and Transformable Access Control scheme for cloud-assisted IoV. Specifically, we leverage a hybrid proxy re-encryption technique to transform ABE ciphertext into identity-based encryption (IBE) ciphertext, making it more suitable for collaborative scenarios. A cuckoo filter is introduced to achieve fully hidden policies, thereby protecting user privacy. Additionally, we incorporate non-interactive zero-knowledge proofs (NIZKPs) to ensure verifiability, guaranteeing the correctness of transformed ciphertexts while preventing malicious accusations against the delegator. A further advantage of our scheme is its support for unbounded attribute spaces, eliminating the need for system reboot when adding new attributes. In addition, our scheme requires only a constant number of pairing operations during the decryption and re-encryption phases, regardless of the number of attributes. We formalize a security model and rigorously prove that our scheme achieves adaptive security. Our scheme achieves simultaneously for the first time cross-primitive ciphertext transformation, fully hidden policy, and verifiability while ensuring adaptive security. We implement our scheme in real-world environments and compare it with existing state-of-the-art schemes. Our approach offers a more comprehensive feature set without significant performance trade-offs, making it highly suitable for cloud-assisted IoV scenarios.
Prof. M. A. Sayyad, Veerendra Yadav, Dr. Geetika M. Patel, Dr. Prakash Deep · 8 authors
The problem of data privacy, interoperability, cyberattacks, and unauthorized changes of sensitive medical records are becoming critical issues in healthcare information exchange systems. The conventional centralized healthcare designs have single-point failures, inadequate transparency, sluggish data synchronization, and insufficient trust management among dispersed medical organizations. In order to overcome these shortcomings, this paper suggests a Blockchain-Assisted Distributed Artificial Intelligence Framework to Secure Healthcare Information Exchange and Data Integrity. The suggested architecture combines a distributed AI-based healthcare analytics system with blockchain-based immutable ledger systems to provide secure, open, and alteration-free medical data exchange among various healthcare nodes. The automated access control and the secure management of authorization is applied using smart contracts, and intelligent anomaly detection and integrity verification of healthcare transactions are implemented using distributed AI modules. The framework also includes encrypted communication and decentralized consensus systems to promote security and reliability in the context of multi-institutional healthcare settings. Simulated healthcare data based on experimentation shows that the proposed framework has a data integrity verification accuracy of 96.4, anomaly detection accuracy of 92.7 and offers both efficient and secure transaction validation performance at a ratio of 41.3 lower than traditional centralized healthcare systems. The suggested architecture enhances the security of healthcare data and trust management, scalability and interoperability of the next-generation intelligent healthcare ecosystems significantly.