Blockchain Papers

Irshad Ullah, Xiaoheng Deng, Xinjun Pei, Husnain Mushtaq · 6 authors

No abstract is available for this record.

Nojan Sheybani, Alessandro Pegoraro, Jonathan Knauer, Phillip Rieger · 7 authors

Split Learning (SL) is a distributed learning approach that enables resource-constrained clients to collaboratively train deep neural networks (DNNs) by offloading most layers to a central server while keeping in- and output layers on the client-side. This setup enables SL to leverage server computation capacities without sharing data, making it highly effective in resource-constrained environments dealing with sensitive data. However, the distributed nature enables malicious clients to manipulate the training process. By sending poisoned intermediate gradients, they can inject backdoors into the shared DNN. Existing defenses are limited by often focusing on server-side protection and introducing additional overhead for the server. A significant challenge for client-side defenses is enforcing malicious clients to correctly execute the defense algorithm. We present ZORRO, a private, verifiable, and robust SL defense scheme. Through our novel design and application of interactive zero-knowledge proofs (ZKPs), clients prove their correct execution of a client-located defense algorithm, resulting in proofs of computational integrity attesting to the benign nature of locally trained DNN portions. Leveraging the frequency representation of model partitions enables ZORRO to conduct an in-depth inspection of the locally trained models in an untrusted environment, ensuring that each client forwards a benign checkpoint to its succeeding client. In our extensive evaluation, covering different model architectures as well as various attack strategies and data scenarios, we show ZORRO's effectiveness, as it reduces the attack success rate to less than 6\% while causing even for models storing \numprint{1000000} parameters on the client-side an overhead of less than 10 seconds.

Tin Tironsakkul, Pradip Kumar Sharma

The advancement in smart mobility communication technology allows intelligent vehicles to interconnect and communicate with each other to improve traffic safety and efficiency. However, the highly dynamic nature of the smart mobility network and vehicle behaviour creates the requirement for effective authentication systems to establish secure and reliable communication between vehicles. The implementation of a reputation system has been proposed to establish trust among untrusted vehicles, where the reliability of a propagated message is assessed based on the reputation of the sender vehicle. Thus, preventing malicious vehicles from potentially broadcasting misleading messages that can cause accidents or disrupt the network. This paper proposes a novel decentralised and dynamic reputation management and computation model based on a consortium blockchain and a multi-signature smart contract. The implementation of blockchain and a smart contract provides a secure and transparent framework for registering vehicles, submitting events, voting feedback, evaluating reputation, and blocking malicious vehicles. To demonstrate the feasibility of the proposed model, we conducted security and performance analyses. The results demonstrate how our model can provide resistance against various attacks, such as data tampering, message forging, self-promotion, vote duplication, bad-mouth, onoff, time-dependent, and collusion attacks.

Dong Fang, Guifang Huang, Shuai Chang, Haonan Yang · 6 authors

Abstract Verifiable encryption enables the decryption to be taken on properly generated ciphertexts, by making the encryptor provide a zero-knowledge proof. To meet the quantum-safe application requirements, such as key escrow, Lyubashevsky et al. proposed a one-shot verifiable encryption (LN17 scheme) based on the hardness of lattice problems. In their scheme, the FSwA-type zero-knowledge proof was obtained using rejection sampling on a discrete Gaussian distribution. In this paper, we present a construction of verifiable encryption that utilizes rejection sampling on bimodal Gaussian to get the associated zero-knowledge proof. Our new construction, while exhibiting a weaker soundness property than LN17 scheme, benefits from a smaller proof size, leading to a reduced size of the verifiable ciphertext. As for the weaker soundness property, it supports some applications such as key escrow where honestly generated verifiable ciphertexts are more useful to be decrypted out in the hope of doing some further computation tasks. We provide the efficiency comparison of the new construction by instantiating it with several sets of concrete parameters.

Siyang Jiang, Hao Yang, Qipeng Xie, Chuan Ma · 8 authors

No abstract is available for this record.

Nicolò Romandini, Carlo Mazzocca, Kai Otsuki, Rebecca Montanari

Blockchain and smart contracts have garnered significant interest in recent years as the foundation of a decentralized, trustless digital ecosystem, thereby eliminating the need for traditional centralized authorities. Despite their central role in powering Web3, their complexity still presents significant barriers for non-expert users. To bridge this gap, Artificial Intelligence (AI)-based agents have emerged as valuable tools for interacting with blockchain environments, supporting a range of tasks, from analyzing on-chain data and optimizing transaction strategies to detecting vulnerabilities within smart contracts. While interest in applying AI to blockchain is growing, the literature still lacks a comprehensive survey that focuses specifically on the intersection with AI agents. Most of the related work only provides general considerations, without focusing on any specific domain. This paper addresses this gap by presenting the first Systematization of Knowledge dedicated to AI-driven systems for blockchain, with a special focus on their security and privacy dimensions, shedding light on their applications, limitations, and future research directions.

Pradyumna Kaushal

Modern vehicles accumulate fragmented lifecycle records across OEMs, owners, and service centers that are difficult to verify and prone to fraud. We propose VehiclePassport, a GAIA-X-aligned digital passport anchored on blockchain with zero-knowledge proofs (ZKPs) for privacy-preserving verification. VehiclePassport immutably commits to manufacturing, telemetry, and service events while enabling selective disclosure via short-lived JWTs and Groth16 proofs. Our open-source reference stack anchors hashes on Polygon zkEVM at <$0.02 per event, validates proofs in <10 ms, and scales to millions of vehicles. This architecture eliminates paper-based KYC, ensures GDPR-compliant traceability, and establishes a trustless foundation for insurance, resale, and regulatory applications in global mobility data markets.

Ehimah Obuse, Noah Ayanbode, Emmanuel Cadet, Iboro Akpan Essien · 5 authors

The convergence of artificial intelligence (AI), multi-access edge computing (MEC), and cloud environments has transformed identity governance by enabling real-time decision-making and seamless access control across decentralized infrastructures. However, this evolution has also introduced complex challenges concerning data privacy, identity trust, and security. This review explores privacy-first security models that integrate AI for identity governance in hybrid cloud-edge architectures. It evaluates privacy-preserving techniques such as homomorphic encryption, federated learning, and zero-knowledge proofs, emphasizing their role in ensuring secure identity authentication, authorization, and auditability. The paper critically analyzes the limitations of conventional identity and access management (IAM) frameworks in dynamic, resource-constrained edge environments and proposes adaptive models that embed privacy by design. Furthermore, the review investigates the interplay between explainable AI (XAI) and policy enforcement for transparent and compliant identity governance. By synthesizing advancements in cryptographic methods, AI reasoning engines, and decentralized identity (DID) systems, the paper outlines a roadmap for building secure, scalable, and privacy-compliant identity infrastructures in the era of pervasive computing. Keywords: Privacy-Preserving Identity Governance, AI-Driven Access Control, Multi-Access Edge Computing (MEC). Federated Identity Management, Explainable AI (XAI), Zero-Knowledge Proofs.

Bharathram Nagaiah -

Blockchain serves as a transformative mechanism for enabling secure, transparent, and privacy-preserving control over data used to train artificial intelligence (AI) models. This paper explores blockchain-enabled frameworks—including data provenance, smart contracts, federated learning integration, Non-Fungible Tokens (NFTs)/DataTokens, and token-based incentive structures—to address data ownership, access governance, contribution compensation, and accountability. We survey platforms such as Ocean Protocol, federated learning with blockchain architectures, and decentralized compute networks. Through analysis of methodologies and case studies across healthcare, IoT, and AI marketplaces, we assess system performance, privacy protection, trust, and regulatory alignment. Our results indicate blockchain facilitates granular data control, immutable provenance, and fair compensation models, yet challenges persist around scalability, incentive fairness, and legal interoperability. We conclude with a roadmap outlining standards, hybrid computations, legal frameworks, and governance models to foster robust "Data-AI-Blockchain" ecosystems.

M Savitha Devi, Ningthoujam Chidananda Singh, Thoudam Basanta Singh

Abstract - Blockchain enabled systems are more and more adopted in healthcare for secured processing of data, but current smart contract usage in healthcare leaks private patient data on execution. The contributions of this paper are two-fold: (1) it proposes a new framework that combines ZKPs with healthcare smart contracts/transactions to achieve full privacy preservation and (2) it discusses the security, usability, and the efficiency of the framework at the same time. Our proposed framework is based on zero-knowledge proof systems zkSNARKs (Zero-Knowledge Succinct Non-Interactive Argument of Knowledge) and zkSTARKs (Zero-Knowledge Scalable Transparent Argument of Knowledge) tailored for computer on medical data without revealing effectively. We conduct extensive analysis and prototype implementation to show that our framework is able to achieve perfect privacy preservation at a 1.87% computational overhead increase with respect to standard smart contracts. The system processes over 10,000 medical records with sub-second verification times and that meet the HIPAA requirements. Experimental results in diverse healthcare applications attest to the efficacy of the approach in practice, and show the substantial gain of privacy preservation (99.8% retention rate) and computational efficiency over the state-of-art algorithms. This paper bridges the gap between blockchain’s transparency and healthcare’s privacy requirements, laying the groundwork for secure and privacy-preserving blockchain based healthcare applications. Key Words: Zero-knowledge proofs, Smart contracts, Healthcare blockchain, Privacy preservation, zkSNARKs, zkSTARKs, Medical data security, HIPAA compliance

Pei Ren, Bo Yang, Yanwei Zhou, Tao Wang · 6 authors

No abstract is available for this record.

Shakeb Ahmed, Waseem Akhtar Khursheed Ahmad, Sajja Suneel, Manpreet Kaur Bhatia · 6 authors

With the increasing need to train AI models on sensitive healthcare data, Federated Learning (FL) has emerged as a decentralized approach that avoids raw data sharing. However, existing methods such as DP-FL and zkFL still suffer from high privacy leakage, computational overhead, and scalability challenges. To overcome these limitations, this study introduces ZK-FedTransformer++, a novel privacy-preserving FL framework. It integrates lightweight TinyViT transformers, zk-SNARKs for verifiable training, differential privacy for statistical protection, and heuristic client selection for robust participation The approach provides secure model updates via cryptographic proof circuits and noise-perturbed gradients. Experiments based on the RSNA Breast Cancer Detection dataset achieve 91.2% accuracy and 35% less privacy leakage. Tools utilized include PyTorch, zk-SNARK libraries, and privacy accounting protocols. In summary, ZK- FedTransformer++ is an effective privacy enhancement, accuracy improvement, and scalability solution that is a feasible solution for secure, decentralized AI applications in real-world healthcare and IoT settings.

Chahna Meka, Keerthi Reddy Palakollu, Azees Maria, Arun Sekar Rajasekaran · 6 authors

No abstract is available for this record.

Michael Herbert Ziegler, Mariusz Nowostawski, Basel Katt

In this literature review, we critically examine the evolving landscape of privacy in blockchain systems, with a particular focus on the differentiation of privacy attacks and protective measures across three distinct layers: the on-chain layer; the off-chain layer; and on the infrastructure, i.e., peer-to-peer network layer. In this review, we categorize prevalent privacy attacks, such as transaction tracing, data leakage, and network surveillance, highlighting their implications at each layer. In addition, we evaluate a range of protective techniques, including cryptographic methods, zero-knowledge proofs, and other privacy-preserving protocols. We explore the compatibility of these privacy techniques with existing blockchain systems. By synthesizing current research and practical implementations, our aims are to provide a comprehensive understanding of privacy challenges and solutions in blockchain environments, identify gaps, and guide future developments in privacy-enhancing technologies within the blockchain ecosystem.

T. Ratha Jeyalakshmi, Alamma Bh, H S Harshitha, K. Agarwal R.

Privacy of users and security of data are important issues that will be exposed to use in the Metaverse by use of Digital Twins (DTs). The current paper suggests a privacy-preserving system, which combines Zero-Knowledge Proofs (ZKPs) of secure identity verification and Federated Learning (FL) of decentralized model training. The framework allows for alleviating the risk of storing data in central facilities and preventing unauthorized access by locally processing data and using cryptographic solutions. The results produced by the evaluation prove that the proposed system is capable of attaining the necessary level of privacy of its users and ensuring reliable and scalable communications within the Metaverse applications.

Mohamad Sheikho Al Jasem, Trevor De Clark, Ajay Kumar Shrestha

The convergence of decentralized artificial intelligence (DAI), blockchain technology, and smart contracts is reshaping the design and governance of intelligent systems. As these technologies rapidly evolve, addressing privacy within their architecture, usage models, and associated risks has become increasingly critical. This systematic literature review examines architectural patterns, governance frameworks, real-world applications, and persistent challenges in DAI systems. It identifies prevailing designs such as federated learning integrated with consensus protocols, smart contract-based incentive mechanisms, and decentralized verification methods. Drawing from a diverse body of recent literature, the review highlights implementations across sectors, including healthcare, finance, IoT, autonomous systems, and intelligent infrastructure, each demonstrating significant contributions to privacy, security, and collaborative innovation. Despite these advancements, DAI systems face ongoing obstacles such as scalability limitations, privacy trade-offs, and difficulties with regulatory compliance. The review emphasizes the need for integrative governance approaches that balance transparency, accountability, incentive alignment, and ethical oversight. These elements are proposed as co-evolving pillars essential to establishing trustworthiness in decentralized AI ecosystems. This work offers a comprehensive review for understanding the current landscape and guiding the development of responsible and effective DAI systems in the Web3 era.

Ahmaddul Hadi, Sandi Rahmadika, Ulfia Rahmi, Harashta Tatimma Larasati · 5 authors

Decentralized technologies such as blockchain and federated learning have emerged as promising solutions to improve privacy, transparency, and security in distributed environments. This paper aims to provide updated research directions concerning the unresolved issues of linkability and traceability in decentralized technology transactions. A systematic review was conducted using Scopus and Web of Science databases, covering studies published between 2017 and 2023. A total of 313 papers were initially identified, screened, and filtered based on inclusion and exclusion criteria, resulting in 29 relevant studies. The analysis indicates that most prior works focused on privacy preservation and incentive mechanisms but neglected linkability and traceability concerns. Several approaches, including ring signatures, CryptoNote protocols, and smart contract-based incentives, were identified as potential solutions. While blockchain–federated learning integration enhances privacy, unresolved traceability and linkability issues still pose significant risks in sensitive domains such as healthcare and finance. Future work should prioritize addressing these issues to ensure secure, anonymous, and scalable decentralized transactions.

Shengnan Zhao, Kuiheng Sun, Chuan Zhao, Bendong Jiang · 6 authors

Machine learning as a service (MLaaS) has emerged as a prominent computing paradigm where users send sensitive data to cloud servers that subsequently return computed results. In MLaaS, ensuring the correctness of these results poses a significant challenge. While zero-knowledge proof (ZKP) presents a potential solution, they often come with substantial memory overhead. Moreover, there is insufficient attention given to the privacy risks associated with untrustworthy servers, which could jeopardize users' sensitive information. In this paper, we introduce$\text{Vp}^{3}\text{CNN}$, a three-party verifiable privacy-preserving convolutional neural network (CNN) inference scheme. In$\text{Vp}^{3}\text{CNN}$, users verify the correctness of CNN inference through a lightweight ZKP protocol grounded in vector oblivious linear evaluation. This protocol is designed to ensure that servers incur minimal memory overhead while maintaining the integrity of the verification process. Based on the optimization of the convolutional relation, the scheme reduces the computational cost associated with the verification process of the convolution operations. In addition,$\text{Vp}^{3}\text{CNN}$employs two non-colluded servers to protect user data privacy via secret sharing schemes. We implement our scheme in C++ and evaluate its performance using the MNIST and CIFAR-10 datasets. Experimental results demonstrate that, compared to existing methods,$\text{Vp}^{3}\text{CNN}$achieves a speedup of 4–5 times in convolution verification while maintaining nearly consistent communication overhead. Importantly,$\text{Vp}^{3}\text{CNN}$does not compromise the accuracy of CNN inference, achieving an accuracy of 97.8% on the MNIST dataset.

Bhabani Sankar Samantray, K. Hemant Kumar Reddy

In today's rapidly evolving landscape of smart city applications, particularly in sensitive areas like the healthcare sector, safeguarding the security, integrity, and privacy of data has become a significant and challenging concern. Specifically in the healthcare sector, the sharing and access of patient records across various stages of care by doctors, nurses, pharmacies, and diagnostic centers introduce new complexities and potential vulnerabilities. However, these challenges intensify more in the case of distributed healthcare networks where data is fragmented across institutions. This work addresses issues such as data vulnerability and misuse in distributed healthcare environments by proposing a Blockchain-enabled Distributed Healthcare System (BeDHS). The model is designed to facilitate secure, transparent, and privacy-preserving collaboration among healthcare entities. It adopts a hybrid approach, integrating a quantum key-based image encryption technique to enhance the security of health records. The encrypted images are securely stored in the InterPlanetary File System (IPFS) to ensure data integrity and availability. Additionally, a Federated Learning (FL) framework is employed to enable collaborative training of AI models across institutions without exposing sensitive patient data. The proposed BeDHS model is implemented using Solidity-based smart contracts on the Ethereum blockchain, ensuring decentralized and tamper-resistant operations. Simulation results demonstrate that the proposed model outperforms existing healthcare data management systems in terms of efficiency and security. • A blockchain-enabled distributed healthcare system is proposed, where the number of healthcare institutions of a smart city are integrated to form a collaborative and transparent model for sharing health records while maintaining security, privacy, and immutability. • A Quantum-Chaos-Encryption cryptographic technique integrated with blockchain for protecting digital documents and medical images from unauthorized access. • To build a privacy-preserved distributed-collaborative healthcare system, a federated learning approach is incorporated that trains the AI models directly at the data source of multiple healthcare institutions while eliminating the need to transfer between the institutions.

Surya B, Karuppasamy L, Selvaragavan S, Yuvan Sankar NKR

The merging of Artificial Intelligence (AI) with the Internet of Things (IoT) has sparked a swift transformation in AIoT systems, allowing for real-time intelligence in smart cities, industries, and homes. Yet, these advancements bring about increasing worries regarding data privacy, device trust, and potential security threats-particularly with the emergence of quantum computing. This paper introduces a secure and privacy focused AIoT framework that integrates Federated Learning with Differential Privacy, Zero-Knowledge Proofs (ZKP) for device authentication, and Post-Quantum Cryptography(CRYSTALSKyber) to protect model updates on the blockchain. Unlike conventional methods that depend on cloud processing and expose sensitive data, this innovative system allows for on-device model training through TinyML, ensuring that data remains on the device. A practical implementation using ESP32-S3 devices in both a smart classroom and home environment showcases the framework's effectiveness. The results indicate a 12% boost in privacy, a 35% reduction in communication costs, and an 8.7% increase in model accuracy compared to traditional methods. This architecture tackles significant unresolved challenges in AIoT by securing data at the edge, preventing device spoofing, and preparing for future quantum threats-making it an excellent choice for privacy-sensitive, real-time AIoT applications.

Jingcheng Zhang, Yekai Zhou, Yingxuan Ren, Man Ho Au · 9 authors

Advancements in sequencing technologies grant individuals unprecedented access to their genomic data. However, existing data management systems or protocols are inadequate in privacy protection, limiting individuals' control over their genomic information, hindering data sharing, and posing challenges for biomedical research. Therefore, demand exists for an owner-governed system fulfilling owner authority, life cycle data encryption, and verifiability simultaneously. Here, we realized Governome, an owner-governed data management system empowering individuals with real-time control over their genomic data. Governome leverages a blockchain to manage transactions and permissions, granting data owners dynamic permission management with full transparency on data usage. It uses homomorphic encryption and zero-knowledge proofs to enable genomic data storage and computation in an encrypted and verifiable form throughout its life cycle. Governome can support versatile genomic applications. We implemented and tested individual variant query, cohort study, genome-wide association study (GWAS) analysis, and forensics on 2,504 1000 Genomes Project (1kGP) genomes, demonstrating its robustness and scalability. Governome is open-source at https://github.com/HKU-BAL/Governome.

Master in Project Management, Md Mohaiminul Hasan

This systematic review examines the role of federated learning (FL) as a privacy-preserving paradigm for enterprise decision systems, synthesizing evidence from 187 peer-reviewed studies. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, the review integrates algorithmic, systems, security, sectoral, and governance perspectives to provide a comprehensive account of current knowledge. Findings highlight that foundational algorithms such as FedAvg, FedProx, and SCAFFOLD dominate the methodological landscape, with significant adaptations emerging to address non-IID and unbalanced datasets across distributed organizational silos. Privacy-preserving mechanisms—including differential privacy, secure aggregation, homomorphic encryption, and multiparty computation—were consistently applied as layered defenses, balancing mathematical guarantees with empirical resilience. The synthesis further revealed critical vulnerabilities to model poisoning, backdoor attacks, and gradient leakage, alongside defensive strategies such as robust aggregation, anomaly detection, and differential privacy clipping. Sector-specific implementations demonstrate FL’s practical utility in healthcare, finance, retail, logistics, telecommunications, and public services, where it enables collaborative modeling without violating data residency or confidentiality requirements. Governance and ethical frameworks, particularly GDPR, CCPA, and the NIST Privacy Framework, were found to shape deployment practices, while documentation artifacts such as datasheets, model cards, and privacy budget ledgers ensure accountability and transparency. Comparative surveys position FL as an integrative socio-technical architecture that unites distributed optimization, privacy engineering, adversarial robustness, and AI governance into a coherent enterprise-ready model. The review concludes that federated learning provides enterprises with a scalable, secure, and ethically aligned approach to leveraging distributed data while preserving trust and compliance.

Gheyath Mustafa Zebari, Nasser Al Musalhi

ABSTRACT As digital transformation accelerates, the demand for secure, decentralized, and intelligent systems continues to rise across industries. Blockchain technology has emerged as a revolutionary tool for ensuring data integrity and transparency, while Artificial Intelligence (AI) offers unparalleled capabilities in enhancing decision‐making, automation, and security. This review explores the intersection of AI and blockchain, focusing on how AI techniques—including machine learning (ML), deep learning (DL), reinforcement learning (RL), and natural language processing (NLP)—can effectively mitigate security vulnerabilities within blockchain ecosystems. By analyzing over 100 peer‐reviewed studies and real‐world applications across finance, healthcare, supply chain management, and smart cities, this paper highlights the strengths and limitations of current AI‐driven solutions for addressing blockchain security challenges. Specifically, it examines AI's role in strengthening consensus mechanisms, detecting anomalies, identifying smart contract vulnerabilities, and preserving data privacy. Additionally, the review identifies emerging trends such as federated learning, quantum‐resistant cryptography, and decentralized AI models. The review concludes by discussing ethical considerations, regulatory challenges, and the necessity for standardized security frameworks to guide the future of AI‐enhanced blockchain security. Ultimately, this review offers actionable insights for building resilient, scalable, and AI‐driven blockchain systems, paving the way for innovation and enhanced security across industries.

Thành Khôi Lê, Thai Le, Hai Anh Tran

A university registration system needs transparency, high security, and high integrity of the data to prevent fraudulent actions on registration. Blockchain technology is a technology focused on the secure, transparent, and integrity of the data. This technology is suitable for the requirements of a registration system. This study proposes a method for integrating the blockchain into the registration system by using the smart contract to design the storage of data on the blockchain, and the functions with the logic process for the actions related to registration. This paper will discuss the architecture design, implementation, and testing to show the potential of applied blockchain in the registration system.