The Internet of Things (IoT) has attained significant interest recently, particularly in the medical field due to the quick development of IoT devices. Medical related data contains a significant volume of personal information, and it is crucial to maintain privacy. As medical information becomes increasingly electronic in the era of big data, securely and accurately storing medical information is critical. However, the heterogeneity of information systems poses a significant challenge to their sharing. Moreover, medical data typically comprises sensitive information, and sharing it can potentially lead to breaches of personal privacy. Data sharing is a significant concern in healthcare because of privacy leakage and security issues. To combat this issue, this paper introduces the prediction and Provable Partitioned Secure Block Chain Principle (PPSBCP) technique is used to secure healthcare data sharing. Initially, in the healthcare data analysis phase, the Preprocessing and normalization are carried out by Z-score normalized for analysing the healthcare-sensitive margins. The SSIR (Sensitive Spectral Impact Rate) method is applied to find the sensitive records. Based on the impact margins, the Binomial Quadratic Sensitive Data Prediction (BQSDP) method is applied to categorize the sensitive and non-sensitive information. In the blockchain phase, create a Hash Index Policy (HIP) to encrypt the data using a Foldable Blockchain Encryption Standard (FBES). The Master Node Handover Authentication Policy (MNHAP) is applied to verify the private key in the data safety. The Distributed Hyper Ledger Mechanism (DHLM) is applied to make the chain transaction principle. The proposed system accomplishes high performance in security by achieving the parameters in verification and validation as well as compared to the existing systems.
Aydin Abadi, Basil Aldali, Benito Vincent, Elliot A. J. Hurley · 13 authors
Weather forecasting plays a vital role in disaster preparedness, agriculture, and resource management, yet current centralized forecasting systems are increasingly strained by security vulnerabilities, limited scalability, and susceptibility to single points of failure. To address these challenges, we propose a decentralized weather forecasting framework that integrates Federated Learning (FL) with blockchain technology. FL enables collaborative model training without exposing sensitive local data; this approach enhances privacy and reduces data transfer overhead. Meanwhile, the Ethereum blockchain ensures transparent and dependable verification of model updates. To further enhance the system's security, we introduce a reputation-based voting mechanism that assesses the trustworthiness of submitted models while utilizing the Interplanetary File System (IPFS) for efficient off-chain storage. Experimental results demonstrate that our approach not only improves forecasting accuracy but also enhances system resilience and scalability, making it a viable candidate for deployment in real-world, security-critical environments.
Secure exchange of patient healthcare data is vital due to the rise of AI in the medical field. However, this advancement introduces challenges such as data breaches, privacy violations, and regulatory demands. Traditional centralized systems store all data in one location, increasing cyberattack risks. This study proposes a secure framework integrating Federated Learning, Blockchain, and Quantum Cryptography. Federated Learning enables decentralized model training without sharing raw data, preserving patient privacy. Blockchain ensures data integrity using an immutable distributed ledger. Quantum Key Distribution (QKD) and AES-256 encryption protect data during transmission and storage. Files are stored in the InterPlanetary File System (IPFS), and their unique Content Identifiers (CIDs) are recorded on the blockchain for tamper-proof verification. Only users with valid quantum-generated keys can decrypt and access the data, ensuring strong privacy and security.
S N Prajwalasimha, Nilesh Shelke, Dilip Kumar Jang Bahadur Saini, Amit Pimpalkar · 6 authors
Federated Learning (FL) is a decentralized collaborative AI training paradigm that maintains privacy of the data. FL is still susceptible to security attacks, malicious clients, and model integrity issues. To mitigate these issues, we introduce a Blockchain-Enabled Federated Learning (BFL) system that incorporates decentralized ledger technology to provide tamper-evident model aggregation, transparent client engagement, and verifiable updates. The suggested BFL framework uses smart contracts to enable automated trust management, zero-knowledge proofs (ZKPs) to facilitate privacy-enhanced authentication, and an incentive mechanism based on tokenized rewards to promote honest engagement. We also propose an adaptive consensus protocol that maximizes blockchain overhead while preserving high scalability for real-world applications like cybersecurity, healthcare, and Industrial IoT (IIoT). Experimental results on benchmark datasets show that BFL dramatically improves model robustness against data poisoning and adversarial attacks with a 15-25% improvement in attack resilience over state-of-the-art FL methods. Our work presents a complete blueprint for secure, privacy-preserving AI and establishes a foundation for the next generation of decentralized intelligence.
This paper proposes a novel architectural framework for robust security within dynamic multi-cloud environments, addressing the limitations of traditional perimeter defenses. It establishes and elaborates upon core Zero-Trust principles, including stringent identity validation, fine-grained access control, and perpetual operational vigilance, to counter contemporary cyber threats such as lateral infiltration and cloud-native attack vectors. The contribution details a systematic approach to fortifying distributed cloud workloads through the enforcement of least-privilege access and micro-segmentation strategies. Furthermore, the paper critically examines advanced policy enforcement mechanisms, enhanced identity management solutions, and the strategic integration of cryptographic and distributed ledger technologies to achieve superior defensive postures. This work delivers actionable insights for designing resilient security postures across diverse cloud infrastructures.
Artificial Intelligence (AI) development in all major fields including healthcare, finance, and intelligent infrastructure raised the need for private, transparent, and secure learning systems. Federated Learning (FL) addresses data privacy by enabling parties to cooperate to train machine learning models without sharing their raw data; however, conventional FL paradigms are based on central aggregators, which introduce vulnerabilities such as single points of failure, decreased transparency, and trust issues among the cooperating parties. This work proposes a novel theoretical framework that integrates blockchain, smart contracts, and Federated Learning to develop a fully decentralized, secure, and auditable AI training platform where smart contracts manage major processes like model aggregation, verification, and reward distribution, disentangling third-party coordination. Blockchain is used as an immutable ledger that openly keeps track of all the updates to the models and participant behavior, enhancing auditability and building trust. The design further incorporates a token- based reward system that will be used to incentivize honest behavior and discourage malicious behavior, addressing root problems of data poisoning and free- riding. This decentralized approach not only improves data confidentiality and system resilience but also promotes fairness and accountability in multi- stakeholder settings. By conceptual analysis and theoretical modeling, the paper lays the foundation for an ethical and scalable AI system that takes advantage of the strengths of Federated Learning, blockchain, and smart contracts, which is a key step towards decentralized intelligence without compromising privacy, integrity, or trust.
Traditional digital identity models suffer from certain vulnerabilities in terms of identity reusability and privacy, as well as a single point of failure. The emergence of the blockchainbased self-sovereign identity model holds promise for addressing these issues in traditional digital identity models. However, existing schemes not only fail to cover privacy preservation throughout the entire lifecycle of credential issuance, verification, and revocation but also present security and efficiency concerns in key rotation. In this paper, we propose a novel blockchain-based self-sovereign identity system and redesign its credential scheme and key rotation mechanism. By leveraging the PS signature and zero-knowledge proof, our scheme preserves the privacy of holders’ private attributes when issuing and verifying credentials. Additionally, with the cryptographic accumulator, our scheme does not reveal any issued or revoked credentials. Furthermore, we propose a secure and efficient key rotation mechanism based on pre-generated key chains, which enables secure and efficient key rotation without relying on a timelock. Finally, we provide a security analysis and performance evaluation, demonstrating the security and practicality of our scheme.
This work presents a cryptographic protocol for secure multi-party verification that achieves com putational privacy while maintaining exceptional computational efficiency. The proposed Position Based Commitment Protocol (PBCP) introduces a position-dependent nonce mechanism combined with cyclic verification architecture, enabling se cure computation over private inputs without re vealing individual parameters. Unlike existing commitment schemes that require complex cryp tographic assumptions, computationally expensive zero-knowledge proofs, or extensive public key in frastructure, Fundamental innovation lies in adapt ing physical laws of fluid dynamics to create nat ural mathematical relationships where each verifi cation equation contains multiple unknowns, mak ing parameter extraction computationally infeasible while preserving verification integrity. The proto col preliminary analysis suggests O(n) communica tion complexity with O(n2) verification complexity, providing substantial improvements over traditional Byzantine Agreement protocols that require O(n3) message exchanges. Comprehensive security analysis reveals robust resistance against statistical attacks with complexity O(R3) where R represents the pa rameter range, complete immunity to timing attacks through blind submission mechanisms, and resilience against collusion attacks involving up to n/2 − 1 ad versarial parties. The protocol’s unique cyclic neigh bor verification creates an interdependent validation network that prevents individual parameter extrac tion while maintaining system-wide integrity through mathematical interdependence rather than crypto graphic assumptions.
Xinyu Zhang, Zerui Chen, Xinhua Cui, Ze Yang · 5 authors
Delegation learning, as a privacy-preserving machine learning paradigm, has been widely applied in large-scale data processing and complex computational tasks in recent years. However, existing schemes still have deficiencies in guarding against malicious behavior and verifying result correctness, and they struggle to balance security and efficiency across different adversary models. In this paper, We propose a rational delegation learning smart contract scheme (SCS-RDL). First, it combines game theory to construct a rational delegation learning framework and introduces a blockchain-based probabilistic verification method. Second, we construct a smart contract scheme and design rational utility functions that effectively incentivize participants’ honest behavior. Finally, experiments demonstrate that SCS-RDL scheme could enhance delegation learning’s training efficiency without sacrificing accuracy and could satisfy public verifiability.
The increase in demand of data driven decision making in sensitive fields like healthcare and finance requires machine learning frameworks that maintain strict data privacy and follow regulations. Federated Learning (FL) provides a decentralized way to train models. It allows multiple organizations to learn together from distributed datasets without sharing raw data. But, traditional FL methods, such as Federated Averaging (FedAvg), face issues in real world situations. These issues arise from different data distributions among clients and the risk of information leaks from shared model updates. In this research study, we introduce a new federated learning framework with two main innovations: First the adaptive aggregation strategy that adjusts client contributions based on how stable they are and their quality, and second an optional differential privacy module at the server to make sure privacy guarantees. We tested the framework on two publicly available datasets: a heart disease dataset from the University of California, Irvine (UCI) repository and a large financial dataset from Kaggle. This simulates collaboration between hospitals and financial institutions. Experimental results show that our adaptive aggregation method boosts model accuracy by up to 4.2% compared to FedAvg, while still performing well even with differential privacy applied. The model achieves an AUC of 0.93 and an F1 score of 0.891, with minimal communication overhead. These results confirm the framework’s strength and its ability to support the ethical use of Artificial Intelligence in regulated and data sensitive areas. They also recommend it can scale effectively across larger federated networks.
Privacy-Preserving Technologies in Data
Artificial Intelligence in Healthcare and Education
Recent years have witnessed a rapid increase in the number of blockchain-based decentralized applications (DApps). As reported by DAppRadar, there are more than 5,000 DApps with more than 17.2 million daily Unique Active Wallets (users). However, it is also reported that some robots are used to manipulate the ranking, attract more users, cheat investors, etc. Hence, it is necessary to detect those robots. Unlike traditional robots or spam detection on Internet, each blockchain has its specific data structure with the impacts of exchanges and whales, leading to the challenges of detecting DApp robots. In this paper, we conduct the first systematic investigation on DApps robots, named DAppCheat. We first collect and release the first multi-blockchain DApp-user dataset, including 4,857 DApps and 99,758,959 users from Ethereum, EOSIO, TRON, and BSC. We propose a general parent account mechanism for multiple blockchains in order to find anonymous user collusion. We define the creating account weight and the used DApp volume weight to reduce the impacts of exchanges and whales. Extensive experimental results show the effectiveness of DAppCheat.
Cloud storage systems have become central to data-driven industries due to their flexibility and scalability. However, ensuring the integrity and confidentiality of outsourced data remains a major concern, particularly in multi-tenant and dynamic cloud environments. This paper proposes a novel privacy-preserving framework that integrates Zero-Knowledge Proofs (ZKP), Pedersen Commitments, and bulk segmentation for efficient and scalable data integrity verification. Unlike traditional approaches, our framework enables Third-Party Auditors (TPAs) to verify cloud-stored data without exposing sensitive information. It is designed to support dynamic operations, detect insider and external threats proactively, and minimize computational overhead through segment-level auditing. Implementation and evaluation using Amazon S3 and DynamoDB demonstrate the framework’s practical viability, low communication cost, and robust tamper detection capabilities.
S N Prajwalasimha, Dilip Kumar Jang Bahadur Saini, Nilesh Shelke, Amit Pimpalkar · 6 authors
Cyber-Physical Systems like smart grids, autonomous cars, and industrial IoT widely implement Federated Learning (FL) to provide distributed intelligence with privacy-protected data. Yet, the impending quantum threat makes conventional cryptographic methods in FL pipelines obsolete, exposing critical infrastructure to future security vulnerabilities. This paper presents Quantum-Resilient Federated Learning (QR-FL), a new framework integrating lattice-based post-quantum cryptography, light-weight zero-knowledge proofs, and trust-aware aggregation ensuring confidentiality, integrity, and quantum/classical attack resistance. Through comprehensive experimentation on real-world CPS datasets, QR-FL provides up to 48% enhanced adversarial robustness, 32% communication overhead savings, and 6.7% enhanced model accuracy compared to current state-of-the-art secure FL solutions. By achieving future-proof security with scalable federated intelligence, QR-FL provides an architecture foundation for future CPS, offering a landmark direction for secure, decentralized AI in the quantum age.
Büşra Büyüktanır, Şahsene Altınkaya, Gozde Karatas Baydoğmus, Kazım Yıldız
Abstract Federated Learning (FL) has emerged as a promising distributed machine learning approach that addresses confidentiality and integrity concerns in various sectors, including Internet of Things (IoT), healthcare, finance, and cybersecurity. In order to improve privacy protection and detection accuracy in decentralized systems, this study investigates the incorporation of FL into Intrusion Detection Systems (IDS). FL is especially useful in situations where data security and privacy are crucial because it allows for the cooperative training of models without centralizing sensitive data. We examine many FL-based IDS solutions across several domains, emphasizing how well they mitigate data breaches, maintain confidentiality, and enhance intrusion detection capabilities. The use of Generative Adversarial Networks (GANs), artificial immune systems, and hybrid deep learning techniques to maximize IDS performance are among the current developments in FL methodology that are covered in the paper. We also look at issues like the requirement for effective aggregation procedures and non-independent and identically distributed (non-IID) data. Finally, we outline future directions and open research topics to improve the scalability, resilience, and effectiveness of FL-based IDS solutions in practical applications.
Purpose: This research addresses critical limitations in existing blockchain-based data sharing solutions by developing an innovative framework integrating zero-knowledge proofs, homomorphic encryption, and smart contract automation for comprehensive big data privacy protection while maintaining utility and regulatory compliance. Methodology: A hierarchical distributed architecture comprising four layers was designed: data owner layer for encryption, blockchain network layer for consensus, privacy protection layer for cryptographic protocols, and application service layer for user interactions. Experimental evaluation was conducted on distributed networks with 20-100 nodes processing$100 ~\text{GB}-5 ~\text{TB}$datasets. Findings: The proposed framework achieves$\text{9 4. 1 \%}$privacy protection strength with$\text{2 2 \%}$computational efficiency improvement compared to existing approaches. The system supports 100 -node deployments while maintaining 131-158 TPS throughput, significantly outperforming traditional zero-knowledge implementations that achieve only 89.3 % privacy strength. Conclusion: The framework represents significant advancement in blockchain-based big data privacy protection, successfully balancing security guarantees with computational efficiency. Practical Implications: The solution demonstrates substantial value for healthcare, financial services, and IoT applications requiring secure collaborative analytics and enterprise-scale data sharing scenarios.
Traditional centralized scholarship evaluation processes typically require students to submit detailed academic records and qualification information, which exposes them to risks of data leakage and misuse, making it difficult to simultaneously ensure privacy protection and transparent auditability. To address these challenges, this paper proposes a scholarship evaluation system based on Decentralized Identity (DID) and Zero-Knowledge Proofs (ZKP). The system aggregates multidimensional ZKPs off-chain, and smart contracts verify compliance with evaluation criteria without revealing raw scores or computational details. Experimental results demonstrate that the proposed solution not only automates the evaluation efficiently but also maximally preserves student privacy and data integrity, offering a practical and trustworthy technical paradigm for higher education scholarship programs.
Gennaro Avitabile, Vincenzo Botta, Daniele Friolo, Ivan Visconti
Balancing immutability and compliance with regulations stands as a significant challenge in the realm of blockchain technology applications. Due to the increase of data-protection requirements (e.g., the GDPR in the EU), it is essential to address the problem of deleting data from a blockchain without compromising the security and transparency of the blockchain itself. Several works proposed techniques to address the data redaction problem. In their seminal work, Ateniese et al. [EuroS&P 2017] were the first to propose a redactable blockchain. Their approach focuses on permissioned blockchains and they showed how to change the content of a transaction without breaking the chaining among blocks by using special cryptographic hash functions (i.e., chameleon hash functions) and secure multi-party computation. We observe that the redaction technique of Ateniese et al. does not take into account the possibility that the blockchain supports smart contracts and that a redaction of a transaction might leave inconsistencies in the logic of the contracts, making some remaining non-redacted transactions invalid, and, more in general, the state of a smart contract inconsistent with the content of transactions. We find this choice rather limiting since decentralized and publicly verifiable computation guaranteed by smart-contract-enabled blockchains is necessary for modern (i.e., Web3) applications. To overcome the above limitations of the applicability of the redaction techniques of Ateniese et al., we propose a redaction technique with wider applicability that leverages succinct non-interactive arguments of knowledge (SNARKs) to realize what we call a proof-of-consistency .
The deployment of artificial intelligence in healthcare is increasingly constrained by privacy, equity, and regulatory compliance challenges, especially in multilingual and cross-border contexts.Traditional centralized machine learning approaches are limited by restrictions on patient data sharing, raising both ethical and legal concerns.Federated learning offers a promising solution by enabling distributed training across institutions without transferring raw data, yet ensuring trust and privacy in federated systems remains a critical barrier.This study proposes a novel framework that combines transformer architectures with encrypted federated datasets anchored by blockchain zero-knowledge proofs (ZKPs) to achieve privacy-preserving, equitable, and multilingual healthcare diagnostics.Transformer-based models, known for their strength in natural language processing and multimodal learning, are adapted to operate on encrypted federated datasets spanning diverse linguistic and demographic contexts.Blockchain provides a decentralized trust layer, while zero-knowledge proofs ensure verifiable model updates without exposing sensitive patient information.This combination allows healthcare providers to collaboratively train diagnostic models that maintain strong predictive performance while adhering to strict privacy guarantees.The framework also advances health equity by enabling multilingual diagnostics that address disparities in underrepresented populations.By integrating explainability mechanisms, stakeholders gain insights into model reasoning across diverse cultural and linguistic datasets.Case applications in federated medical imaging, multilingual clinical notes, and genomic diagnostics highlight the framework's capacity to balance accuracy, privacy, and fairness.Overall, the integration of transformers, federated learning, and blockchain ZKPs represents a pathway toward trustworthy and equitable AI-driven healthcare, enabling collaborative innovation while safeguarding patient rights.
With the widespread adoption of Internet of Things (IoT) technologies in healthcare systems, security issues related to user privacy during data transmission and sharing have become increasingly prominent. To address these challenges, this paper proposes a medical privacy protection and secure sharing scheme based on Quantum Key Distribution (QKD). The scheme integrates multiple technologies, including blockchain, smart contracts, zero-knowledge proofs, and Chebyshev chaotic mapping, to ensure secure data sharing and access control among multiple communication entities. Compared with existing solutions, our approach enhances key management security through quantum keys and improves communication resilience against attacks by leveraging chaotic systems. User identity privacy is protected via zero-knowledge proofs. Under the random oracle model, the security of the proposed scheme is formally proven. Moreover, comparative experiments with existing protocols demonstrate the scheme's comprehensive advantages in terms of security and performance, evaluated across throughput, computational overhead, communication overhead, and storage overhead.
Jiacheng Yang, Yongxin Zhang, Hong Lei, Zijian Bao · 6 authors
In the process of integrating the digital economy with the real economy, a vast and diverse supply of data has emerged. Among these, the exponential growth of data in vehicular ad-hoc networks (VANETs) hold immense commercial value. This further drives the demand for building large-scale data marketing platforms to support trading between vehicles and businesses in order to reduce the cost of local management. However, this must address several challenges related to security and performance, such as fairness, privacy protection, and data delivery efficiency. Therefore, this paper proposes a privacy-preserving large-scale data marketing system (PLDM), aiming to address these challenges. Specifically, this solution is based on blockchain to build a decentralized trusted third party to ensure the fairness of the trading process. In addition, we combine the$\Sigma$-protocol and Merkle tree to prove the validity of both the encryption of data to be traded and the identities of the trading participants. This not only achieves privacy protection for data and identities but also reduces the computational costs for vehicles. We provide the security analysis and experimental evaluation ofPLDM. And the results show thatPLDMperforms well in fairness and privacy protection, supporting efficient delivery of large-scale data and low on-chain computational costs.
Rayhan Ferdous Srejon, M. Fahim, Sk. Md. Shadman Ifaz, Saha Reno · 5 authors
Traditional ride-sharing platforms rely on centralized architectures, leading to concerns about data privacy, transparency, and security vulnerabilities. These issues, coupled with high service fees and susceptibility to cyber threats, highlight the need for a decentralized, privacy-preserving alternative. To address these challenges, we propose a semi-public blockchain-based ride-sharing platform that leverages Hyperledger Fabric for secure, permissioned data storage and IPFS for distributed data management. Our system integrates Ethereum smart contracts for transaction transparency and incorporates the Cosmos SDK to enable seamless interoperability between private and public blockchains. Additionally, we introduce a fair “pay-as-you-drive” mechanism, replacing conventional time-locked deposit protocols to ensure users only pay for the distance traveled. By combining permissioned blockchain security with decentralized storage and efficient cross-chain communication, our approach ensures privacy, scalability, and real-world applicability. Our performance analysis demonstrate that our Proof-of-Stake consensus mechanism achieves 15–20 percent higher throughput and 30–40 percent lower latency compared to PoW and PoA, alongside reduced CPU usage under high transaction loads. With a monthly operational cost of approximately 30,000 BDT per node, the platform proves both affordable and scalable, offering a practical path toward decentralized, secure ride-sharing solutions.