Background: Even while traditional Raft is effective at leader election and log replication, it is not appropriate for sensitive applications like supply chains, financial systems, or healthcare because it lacks built-in privacy safeguards. Materials and Methods: A privacy-preserving Raft consensus method is proposed to solve the privacy issues that occur when private information is transferred between nodes in a distributed system such as a blockchain. Raft itself, by default, does not provide any steps toward ensuring data confidentiality during consensus. By employing privacy-preserving cryptographic techniques like homomorphic encryption and zero-knowledge proofs, nodes can reach consensus while keeping sensitive data private. Results: Traditional Raft performs much better in scenarios where performance matters, while Privacy-Perving Raft works better in a sensitive application to privacy (the average of write throughput is 5% lower than that of traditional Raft) and CPU is 40-60%. Conclusion: Based on the gained privacy by some computational costs, it will be valid to draw the conclusion that this works for privacy-sensitive applications within decentralized systems with these performance and security analyses.
In response to the issues of high transaction transparency and regulatory difficulties in blockchain account-model transactions, this paper presents a supervised blockchain anonymous transaction model based on certificateless signcryption aimed at ensuring secure blockchain transactions while minimizing both computational and communication overhead. During the transaction process, this approach utilizes certificateless public key signcryption without bilinear pairs to generate anonymous user identities, achieving strong anonymity of user identities and confidentiality of transaction amounts. It employs the Paillier homomorphic encryption algorithm to update transaction amounts and uses the FO commitment-based zero-knowledge proof scheme to validate transaction legality. Additionally, adopting a publicly verifiable secret threshold sharing scheme for hierarchical regulatory authority reduces the security risk of a single regulator storing the regulatory key. This model not only meets the privacy and timely update requirements of account-based blockchain transactions but also effectively regulates abnormal transactions. Rigorous security analysis and proofs demonstrate that this model possesses excellent anonymity, traceability, forward security, and backward security. When compared to similar schemes, the computational cost is reduced by at least 33.18%, effectively fulfilling the requirements for security.
The exponential rise of healthcare data calls for strong privacy-preserving methods for collaborative learning. In this paper, an Integrated Federated Learning Framework integrating homomorphic encryption, blockchain, and adaptive learning approaches is introduced to meet the challenges of privacy, scalability, and heterogeneity of data in healthcare. Data confidentiality is provided by homomorphic encryption with the ability to carry out computations over encrypted data safely, and tamper-proof sharing of data by blockchain. Adaptive learning mechanisms such as weighted federated learning and SMOTE address data imbalance and heterogeneity. Evaluated on real-world data such as MIMIC-III, the framework achieves 93.2% model accuracy, outperforming traditional centralized and federated learning approaches. Experimental results are a 30% reduction in communication overhead with faster convergence, making it feasible for large-scale, privacy-sensitive healthcare applications. The proposed framework provides opportunities for secure, efficient, and collaborative analysis of healthcare data.
Blockchain technology, a decentralized and immutable ledger, has transformed identity and access management (IAM) by enhancing security, privacy, and trust in digital ecosystems. Ensuring safe authentication and data integrity is made possible by its integration with sophisticated cryptographic techniques like zero-knowledge proofs (ZKPs) and public- key infrastructure (PKI). Other methods include verifiable credentials (VCs) and decentralized identifiers (DIDs). This paper provides a comprehensive analysis of blockchain-based IAM systems, comparing leading blockchain platforms, including Ethereum, Hyperledger Indy, IOTA, and IoTeX, in identity management. The role of blockchain in mitigating identity-related threats, such as identity theft and unauthorized access, is explored through decentralization, immutability, and smart contract automation. Additionally, key security enhancements, including cryptographic mechanisms that strengthen decentralized identity solutions and privacy-preserving authentication, are examined. The potential of blockchain to establish a self-sovereign identity framework that fosters trust, scalability, and security in digital identity ecosystems is highlighted, paving the way for the next generation of identity management solutions.
Aishwarya Parab, P. Pradhan, Yogesh Simmhan, Arnab K. Paul
The increasing availability of data from diverse sources, including trusted entities such as governments, as well as untrusted crowd-sourced contributors, demands a secure and trustworthy environment for storage and retrieval. Blockchain, as a distributed and immutable ledger, offers a promising solution to address these challenges. This short paper studies the feasibility of a blockchain-based framework for secure data storage and retrieval across trusted and untrusted sources, focusing on provenance, storage mechanisms, and smart contract security. Through initial experiments using Hyper Ledger Fabric (HLF), we evaluate the storage efficiency, scalability, and feasibility of the proposed approach. This study serves as a motivation for future research to develop a comprehensive blockchain-based storage and retrieval framework.
With the emergence of illegal behaviors such as money laundering and extortion, the regulation of privacy-preserving cryptocurrency has become increasingly important. However, existing regulated privacy-preserving cryptocurrencies usually rely on a single regulator, which seriously threatens users’ privacy once the regulator is corrupt. To address this issue, we propose a linkable group signature against malicious regulators (ALGS) for regulated privacy-preserving cryptocurrencies. Specifically, a set of regulators work together to regulate users’ behavior during cryptocurrencies transactions. Even if a certain number of regulators are corrupted, our scheme still ensures the identity security of a legal user. Meanwhile, our scheme can prevent double-spending during cryptocurrency transactions. We first propose the model of ALGS and define its security properties. Then, we present a concrete construction of ALGS, which provides CCA-2 anonymity, traceability, non-frameability, and linkability. We finally evaluate our ALGS scheme and report its advantages by comparing other schemes. The implementation result shows that the runtime of our signature algorithm is reduced by 17% compared to Emura et al. (2017) and 49% compared to KSS19 (Krenn et al. 2019), while the verification time is reduced by 31% compared to Emura et al. and 47% compared to KSS19.
Hao-Tse Chung, Shao‐Hung Cheng, Yu‐Jia Chen, Li‐Chun Wang
Emerging Blockchain-empowered Federated Learning (BCFL) technology combines the decentralized security of blockchain with the privacy protection of federated learning. BCFL addresses the issue of single points of failure in centralized systems, making it an increasingly popular solution. However, current consensus mechanisms, such as Proof of Work (PoW), Proof of Stake (PoS), and Practical Byzantine Fault Tolerance (PBFT), lead to challenges such as high computational costs and limited scalability. This paper proposes a Batch Zero-Knowledge Proof-based practical Byzantine fault-tolerant (BZ-BFT) consensus mechanism for BCFL to enhance efficiency and reliability. By integrating Zero-Knowledge Proof (ZKP), our approach enables the verification of the primary node's proposal without revealing information from other network nodes, thereby ensuring the credibility of the aggregated results. To address the high computational overhead associated with ZKP, we present a batch quantization preprocessing technique called BatchZKP. Our proposed BZ-BFT reduces initialization, proof generation, and verification time by$97.81 \%, 70.0 \%$, and 47.64 %, respectively, significantly boosting BCFL system efficiency and reliability. Additionally, our approach reduces communication complexity from$O\left(n^{2}\right)$to$O(n)$and enhances Byzantine fault tolerance to${1/2}$.
Muhammad Firdaus, Harashta Tatimma Larasati, Kyung Hyune-Rhee
Healthcare data is often fragmented across various institutions due to its highly sensitive and private nature. In this sense, hospitals and clinics maintain electronic health records (EHRs) independently; hence, valuable data is siloed within individual organizations, preventing comprehensive analysis that could benefit from diverse data sources. Federated learning (FL) addresses these challenges by enabling the training of a shared global model using data distributed across multiple institutions without moving the data from its source. By leveraging FL, healthcare institutions can combine their data assets to improve predictive analytics, personalized medicine, and overall healthcare outcomes, ultimately benefiting patients and the healthcare system. However, the current FL model with a central server presents several challenges within healthcare, including the risk of malicious attacks, regulatory compliance, and privacy vulnerabilities. To overcome these issues, this paper introduces the FL framework with blockchain and homomorphic encryption (HE). Our framework aims to minimize the role of the central server, enable collaborative model training across healthcare organizations, and enhance data security and privacy. In this sense, blockchain ensures the integrity and transparency of the process, while homomorphic encryption ensures that the data remains private. This framework can potentially enable institutions to enrich medical knowledge while securely keeping patient data collaboratively and facilitating healthcare analytics in practical settings.
A Decentralized Identifier (DID) empowers an entity to prove control over a unique and self-issued identifier without relying on any identity provider. The public key material for the proof is encoded into an associated DID document (DDO). This is preferable shared via a distributed ledger because it guarantees algorithmically that everyone has access to the latest state of any tamper-proof DDO but only the entities in control of a DID are able to update theirs. Yet, it is possible to grant deputies the authority to update the DDO on behalf of the DID owner. However, the DID specification leaves largely open on how authorizations over a DDO are managed and enforced among multiple deputies. This article investigates what it means to govern a DID and discusses various forms of how a DID can be controlled by potentially more than one entity. It also presents a prototype of a DID-conform identifier management system where a selected set of governance policies are deployed as Smart Contracts. The article highlights the critical role of governance for the trustworthy and flexible deployment of ledger-anchored DIDs across various domains.
In an era of extensive data collection, preserving individual privacy while deriving actionable insights is a critical challenge. This study proposes a Privacy-Preserving Skellam Mixture Matrix Fusion (PPS MMF) framework enhanced by a Bidirectional Encoder Representations from Transformers (BERT) model. The PPS MMF leverages the Skellam distribution to obscure sensitive information during data fusion, ensuring privacy preservation. By integrating BERT, the framework captures nuanced contextual information, improving the accuracy of downstream tasks. Operating in a decentralized manner, this approach mitigates centralized data breach risks and enables secure data fusion across domains like healthcare, finance, and social media. Experimental results validate its effectiveness and practicality in real-world applications.
Machine Learning (ML) in distributed environments increasingly deals with sensitive data (like healthcare or financial records) that cannot be centrally stored or processed due to privacy concerns. Federated Learning (FL) addresses this by enabling model training across decentralized devices, but faces significant challenges including system reliability, node failures, and trust issues among participants. Traditional FL approaches often rely on centralized coordinators, creating single points of failure and potential security vulnerabilities. This paper presents a novel approach to FL that leverages smart contracts, blockchain, and decentralized storage to enhance the traceability and reliability of the learning process. Our proposed system architecture is fully decentralized, eliminating single points of failure and promoting cooperation through a rewarding mechanism. Unlike previous approaches that neglect node fault tolerance, we introduce a smart contract based scheme for managing node failures and electing the aggregator node. The presence of the smart contract, executed on a decentralized permissioned blockchain, provides reliability guarantees and eliminates the need for costly distributed algorithms in terms of message exchange. An experimental study is conducted to evaluate various aspects of the FL system. We present results related to the accuracy and effectiveness of the FL system on ML models. We also examine the performance related to the distribution of the weights of the ML model based on the use of IPFS. Furthermore, we analyze the performance of the smart contract in terms of gas consumption. Lastly, we investigate the impact of failures combined with incentive policies and aggregator election algorithms on the FL system. Our findings demonstrate the viability of the proposed approach, paving the way for more robust, reliable, and efficient FL systems.
N.V.S.L. Ganesh, Marripati Tharun Kumar Reddy, G. Sabeena Gnana Selvi
Privacy-Preserving Machine Learning (PPML) is essential for securing sensitive data during model training and deployment. This paper proposes an integrated framework combining Federated Learning (FL), Differential Privacy (DP), Secure Multi-Party Computation (SMPC), and Homomorphic Encryption (HE) to enhance data privacy while maintaining model utility. FL enables decentralized training without data sharing, DP ensures individual data anonymity, SMPC facilitates collaborative learning without exposing private data, and HE allows computations on encrypted data. We analyze the computational complexity, privacy guarantees, and model effectiveness through empirical evaluation. Results demonstrate that FL and DP achieve an optimal balance of privacy and accuracy, while HE ensures maximum security at a higher computational cost. Privacy audits confirm strong resilience against membership inference and model inversion attacks. The study also highlights practical limitations and trade-offs, offering insights into scalable and efficient deployment of PPML techniques in healthcare, finance, and IoT applications.
Marko Štaka, Miroslav Stefanović, Darko Stefanović, Đorđe Pržulj · 5 authors
Information systems in the healthcare sector face significant challenges regarding the protection of user data. These systems process highly sensitive data, including medical and private information of patients. Unauthorized access and breaches of data security are among the most commonly identified issues in data protection. Inadequate security of information systems negatively impacts the integrity of healthcare systems and jeopardizes the safety of citizens who use these services. To address these problems, the implementation of smart contracts is proposed, which would regulate access rights to important data. Leveraging the potential of smart contracts can contribute to reducing security risks and enhancing system security. This paper discusses the technical, legal, and ethical issues related to the implementation of smart contracts in healthcare systems, along with their benefits and challenges.
Federated Learning (FL) has emerged as a promising paradigm in distributed machine learning, enabling collaborative model training while preserving data privacy. However, despite its many advantages, FL still contends with significant challenges -- most notably regarding security and trust. Zero-Knowledge Proofs (ZKPs) offer a potential solution by establishing trust and enhancing system integrity throughout the FL process. Although several studies have explored ZKP-based FL (ZK-FL), a systematic framework and comprehensive analysis are still lacking. This article makes two key contributions. First, we propose a structured ZK-FL framework that categorizes and analyzes the technical roles of ZKPs across various FL stages and tasks. Second, we introduce a novel algorithm, Verifiable Client Selection FL (Veri-CS-FL), which employs ZKPs to refine the client selection process. In Veri-CS-FL, participating clients generate verifiable proofs for the performance metrics of their local models and submit these concise proofs to the server for efficient verification. The server then selects clients with high-quality local models for uploading, subsequently aggregating the contributions from these selected clients. By integrating ZKPs, Veri-CS-FL not only ensures the accuracy of performance metrics but also fortifies trust among participants while enhancing the overall efficiency and security of FL systems.
The integration of Internet of Things (IoT) devices in healthcare has enhanced medical efficiency but poses challenges such as data privacy risks and internal abuse. Traditional IoT data access frameworks suffer from centralization, limited scalability, and static permission controls. To address these issues, we propose MedAccessX, a blockchain-based access control framework combining attribute-based access control (ABAC) and role-based access control (RBAC). MedAccessX utilizes four types of smart contracts: a user management contract (UMC) for managing user operations, a medical data management contract (MDMC) for handling data, a policy contract (PC) for managing access rights, and an access control contract (ACC) for enforcing permissions and facilitating data sharing. Our evaluation, conducted on a private Ethereum blockchain network with multiple nodes, assesses security, deployment cost, gas consumption, throughput, and response time. Comparative analysis demonstrates that MedAccessX achieves lower deployment costs and higher throughput, outperforming existing solutions.
Federated learning (FL) is an emerging paradigm that enables multiple clients to collaboratively train a machine learning (ML) model without the need to exchange their raw data. However, it relies on a centralized authority to coordinate participants’ activities. This not only interrupts the entire training task in case of a single point of failure, but also lacks an effective regulatory mechanism to prevent malicious behavior. Although blockchain, with its decentralized architecture and data immutability, has significantly advanced the development of FL, it still struggles to withstand poisoning attacks and faces limitations in computational scalability. We propose Zkfhed, a verifiable and scalable FL system that overcomes the limitations of blockchain-based FL in poison attacks and computational scalability. First, we propose a two-stage audit scheme based on zero-knowledge proofs (ZKPs), which verifies that the training data are extracted from trusted organizations and that computations on the data exactly follow the specified training protocols. Second, we propose a homomorphic encryption delegation learning (HEDL), based on fully homomorphic encryption (FHE). It is capable of outsourcing complex computing to external computing resources without sacrificing the client's data privacy. Final, extensive experiments on real-world datasets demonstrate that Zkfhed can effectively identify malicious clients and is highly efficient and scalable in terms of online time and communication efficiency.
With the development of communication infrastructure and the popularity of smart devices, e-commerce is presenting in more diverse forms and attracting the attention of more and more users. Since e-commerce transactions usually involve sensitive information of a large number of users, privacy and security have become increasingly important issues. Despite certain advantages (e.g., trading security), the privacy protection capability and efficiency of blockchain is still limited by some key factors, especially of its architecture. In this paper, we propose a blockchain-based privacy protection system named PBTMS that integrates zero-knowledge proofs, hybrid encryption, and Pedersen commitments as foundational mechanisms to ensure robust privacy protection for transaction data and user information. To achieve secure, reliable, and efficient e-commerce transactions, the PBTMS employs blockchain technology and consensus mechanisms to enable distributed storage, thereby mitigating single points of failure and addressing the risks posed by malicious nodes. Moreover, by integrating on-chain storage with off-chain computation, the system substantially reduces blockchain-related overheads, including processing time, gas consumption, and storage costs. This design establishes the PBTMS as a highly adaptable and efficient system for the evolving requirements of secure and privacy-preserving e-commerce platforms. Theoretical analysis and experimental validation demonstrate that PBTMS reduces decryption and authentication times by 79.2% and 52.6%, respectively, while cutting encrypted data size by 52.5% and overall gas consumption by 55.4%, outperforming state-of-the-art solutions. These results indicate that PBTMS is a reliable and efficient system for secure e-commerce transaction platforms and provides a novel approach to enhancing privacy protection in e-commerce.
As artificial intelligence (AI) becomes integral to microservices deployed across multi-cloud environments, ensuring secure and scalable observability is critical. Traditional centralized observability methods often fail to address the privacy, compliance, and performance challenges inherent to distributed AI systems. This paper presents a federated learning–based framework for AI observability that preserves data privacy and scalability across heterogeneous cloud platforms. The proposed framework decentralizes telemetry collection and analysis by integrating local observability agents with secure federated aggregation, while maintaining interoperability with modern DevOps pipelines. We evaluate the architecture through case studies in retail, healthcare, and finance sectors, demonstrating improvements in anomaly detection, regulatory compliance, and operational efficiency. Additionally, the paper examines ethical considerations such as data privacy, fairness, and transparency, and outlines future directions including edge observability, privacy-enhanced computation, and automated governance. This research provides a foundational strategy for building trustworthy and efficient observability systems tailored to AI-powered microservices within complex multi-cloud ecosystems. Traditional observability methods struggle with privacy and performance in AI-powered multi-cloud microservices. We propose a federated learning–based framework that enables decentralized telemetry monitoring while ensuring compliance and scalability. Our evaluation across healthcare, finance, and retail shows improvements in anomaly detection latency (25%), fraud detection accuracy (18%), and GDPR/HIPAA alignment. This work lays the groundwork for trustworthy and efficient AI observability in complex cloud-native ecosystems.
Property transactions in the UK are increasingly adopting blockchain technology to enhance efficiency, transparency, and security. However, the inherent transparency of blockchain raises significant data privacy risks and regulatory compliance challenges, particularly under the UK General Data Protection Regulation (UK GDPR). This study examines the role of Zero-Knowledge Proofs (ZKPs) in addressing these concerns by enabling transaction validation while preserving confidentiality. Using entropy measures, k-anonymity analysis, and logistic regression, this research quantitatively assesses the privacy risks, effectiveness of ZKPs, and regulatory acceptance in blockchain-based property transactions. The findings reveal that 65.5% of transactions remain highly or moderately identifiable, posing privacy vulnerabilities under UK data protection laws. ZKP-enabled transactions significantly enhance confidentiality, achieving a 92.5% transaction privacy score, compared to 48.3% for non-ZKP transactions. However, these privacy gains come at a 67.8% increase in transaction costs, highlighting a critical trade-off between security and efficiency. Regulatory approval rates for ZKP-based blockchain platforms stand at 72.5%, suggesting a strong potential for compliance advantages. While ZKPs improve privacy and regulatory alignment, challenges remain in terms of computational overhead, transaction costs, and adoption barriers. To facilitate large-scale implementation, this study recommends optimizing zk-Rollups for efficiency, developing clear policy frameworks, and enhancing collaboration between regulators, industry stakeholders, and blockchain developers. These steps are essential to ensuring a balance between privacy, scalability, and compliance, paving the way for secure and legally sound blockchain-based property transactions in the UK.
The arrival of Machine Learning (ML) completely changed how we can unlock valuable information from data. Traditional methods, where everything was stored in one place, had big problems with keeping information private, handling large amounts of data, and avoiding unfair advantages. Machine Learning has become a powerful tool that uses Artificial Intelligence (AI) to overcome these challenges. We started by learning the basics of Machine Learning, including the different types like supervised, unsupervised, and reinforcement learning. We also explored the important steps involved, such as preparing the data, choosing the right model, training it, and then checking its performance. Next, we examined some key challenges in Machine Learning, such as models learning too much from specific examples (overfitting), not learning enough (underfitting), and reflecting biases in the data used. Moving beyond centralized systems, we looked at decentralized Machine Learning and its benefits, like keeping data private, getting answers faster, and using a wider variety of data sources. We then focused on a specific type called federated learning, where models are trained without directly sharing sensitive information. Real-world examples from healthcare and finance were used to show how collaborative Machine Learning can solve important problems while still protecting information security. Finally, we discussed challenges like communication efficiency, dealing with different types of data, and security. We also explored using a Zero Trust framework, which provides an extra layer of protection for collaborative Machine Learning systems. This approach is paving the way for a bright future for this groundbreaking technology.