One revolutionary way to tackle privacy and security issues in federated learning (FL) is to include blockchain technology and zero-knowledge proofs (ZK) into machine learning frameworks. To strengthen FL's defences against threats such as model poisoning attacks, this work investigates the use of ZK proofs. This study presents a new technique that uses secure multi-party computation (MPC) to efficiently detect poisoned models, addressing the shortcomings of previous ZK systems. Data anonymization, encryption of sensitive information, and encoding of categorical data all contribute to the proposed model's privacy-preserving features. Adding a privacy-protecting layer is an integral part of ML model integration. ZK circuits employ ZK-SNARKs or Bulletproofs to generate proofs that the ML model may use to predict without disclosing the data. ZK-SNARKs are trusted, and request validation and data access rules control proof access.
In this paper, we propose a Proof-of-Location (PoL)-based location verification scheme for mitigating Sybil attacks in vehicular ad hoc networks (VANETs). For this purpose, we employ smart contracts for storing the location information of the vehicles. This smart contract is maintained by Road Side Units (RSUs) and acts as a ground truth for verifying the position information of the neighboring vehicles. To avoid the storage of fake location information inside the smart contract, vehicles need to solve unique computational puzzles generated by the neighboring RSUs in a limited time frame whenever they need to report their location information. Assuming a vehicle has a single Central Processing Unit (CPU) and parallel processing is not allowed, it can solve a single computational puzzle in a given time period. With this approach, the vehicles with multiple fake identities are prevented from solving multiple puzzles at a time. In this way, we can mitigate a Sybil attack and avoid the storage of fake location information in a smart contract table. Furthermore, the RSUs maintain a dedicated blockchain for storing the location information of neighboring vehicles. They take part in mining for the purpose of storing the smart contract table in the blockchain. This scheme guarantees the privacy of the vehicles, which is achieved with the help of a PoL privacy preservation mechanism. The verifier can verify the locations of the vehicles without revealing their privacy. Experimental results show that the proposed mechanism is effective in mitigating Sybil attacks in VANET. According to the experiment results, our proposed scheme provides a lower fake location registration probability, i.e., lower than 10%, compared to other existing approaches.
Rafiqul Islam Munna, Kazi Md. Rokibul Alam, Yasuhiko Morimoto
Data sharing across collaborative mining can aid the community through analyses for decision-making tasks. While sharing personal data with a 3rdparty, data anonymization is a lawful obligation. Besides, preventing unauthorized access is another demand for storing data in a database. This paper proposes a framework of location data anonymization and data privacy over the database to ensure utility while data mining. For anonymization, upon the location data, it exploits dynamic geofencing, gridding, differential privacy, and zero-knowledge proof consecutively to perform the geographical analyses. Also, to securely store the data in the database, it encrypts data by elliptic curve cryptography. The anonymization framework produces grid-based circular coordinate boundaries, adds controlled noise, and proves the 3rdparty about its query results without telling any knowledge of the location (i.e., coordinates). Finally, the performance evaluation, analyses, comparisons, etc., using real data demonstrate that the proposed framework retains a better trade-off between the data utility and anonymity than the state-of-the-art works.
This study focuses on data encryption and privacy protection technologies in cloud computing environments. By systematically implementing and evaluating various encryption algorithms (such as AES, RSA, and homomorphic encryption) and privacy protection techniques (including data masking, differential privacy, secure multi-party computation, and zero-knowledge proofs), the feasibility and effectiveness of these technologies in cloud environments are explored. A simulated cloud environment was constructed for experiments, and the results indicate that AES performs excellently in large-scale data processing, while homomorphic encryption demonstrates unique advantages in specific scenarios. Privacy protection techniques can achieve a balance between protecting user privacy and maintaining data availability. System performance and security tests confirm that the proposed solutions effectively support the data security and privacy protection needs in large-scale cloud environments. This research provides a comprehensive technical implementation and evaluation reference for data security and privacy protection in cloud computing environments, while also highlighting some challenges and offering valuable insights for future research directions.
Frequent user data breaches and misuse incidents highlight the flaws in current identity management systems. This study proposes a blockchain-based, peer-supervised self-sovereign identity (SSI) generation and privacy protection technology. Our approach creates unique digital identities on the blockchain, enabling secure cross-domain recognition and data sharing and satisfying the essential users' requirements for SSI. Compared to existing SSI solutions, our approach has the practical advantages of less implementation cost, ease of users' understanding and agreement, and better possibility of being soon adopted by current society and legal systems. The key innovative technical features include (1) using a zero-knowledge proof technology to ensure data remain "usable but invisible", mitigating data breach risks; (2) introducing a peer review mechanism among service providers to prevent excessive data requests and misuse; and (3) implementing a comprehensive multi-party supervision system to audit all involved parties and prevent misconduct.
The swift progression of Web3 and the proliferation of Decentralized Applications (DApps) have ushered in an era where data services are seamlessly integrated with blockchain technology. Despite this integration, the highly esteemed Quality of Service (QoS) service recommendation methodologies from the Web2.0 era face challenges in achieving seamless compatibility due to their centralized nature. In this paper, we introduce an innovative hybrid approach that bridges the on-chain and off-chain realms for service evaluation and recommendation, which we term as QoBS. This method leverages the power of ring signatures to safeguard identity data, thereby ensuring an efficient and secure framework for decentralized blockchain governance in the Web3 ecosystem. Through rigorous experimentation within the Ethereum environment, we validate the practical viability of our proposed solution, showcasing its robustness and effectiveness in the evolving landscape of decentralized services.
Jianping Yu, Hang Yao, Kai Ouyang, Xiaojun Cao · 5 authors
Federated Learning (FL) enables clients to securely share gradients computed on their local data with the server, thereby eliminating the necessity to directly expose their sensitive local datasets. In traditional FL, the server might take advantage of its dominant position during the model aggregation process to infer sensitive information from the shared gradients of the clients. At the same time, malicious clients may submit forged and malicious gradients during model training. Such behavior not only compromises the integrity of the global model, but also diminishes the usability and reliability of trained models. To effectively address such privacy and security attack issues, this work proposes a Blockchain-based Privacy-preserving and Secure Federated Learning (BPS-FL) scheme, which employs the threshold homomorphic encryption to protect the local gradients of clients. To resist malicious gradient attacks, we design a Byzantine-robust aggregation protocol for BPS-FL to realize the cipher-text level secure model aggregation. Moreover, we use a blockchain as the underlying distributed architecture to record all learning processes, which ensures the immutability and traceability of the data. Our extensive security analysis and numerical evaluation demonstrate that BPS-FL satisfies the privacy requirements and can effectively defend against poisoning attacks.
Recommender systems (RS) play an integral role in many online platforms. Exponential growth and potential commercial interests are raising significant concerns around privacy, security, fairness, and overall responsibility. The existing literature around responsible recommendation services is diverse and multidisciplinary. Most literature reviews cover a specific aspect or a single technology for responsible behavior, such as federated learning or blockchain. This study integrates relevant concepts across disciplines to provide a broader representation of the landscape. We review the latest advancements toward building privacy-preserved and responsible recommendation services for the e-commerce industry. The survey summarizes recent, high-impact works on diverse aspects and technologies that ensure responsible behavior in RS through an interconnected taxonomy. We contextualize potential privacy threats, practical significance, industrial expectations, and research remedies. From the technical viewpoint, we analyze conventional privacy defenses and provide an overview of emerging technologies including differential privacy, federated learning, and blockchain. The methods and concepts across technologies are linked based on their objectives, challenges, and future directions. In addition, we also develop an open source repository that summarizes a wide range of evaluation benchmarks, codebases, and toolkits to aid the further research. The survey offers a holistic perspective on this rapidly evolving landscape by synthesizing insights from both RS and responsible AI literature.
The Borda Count method, a widely used ranked voting system, is known for its fairness and simplicity. However, when applied to large-scale voting systems, it faces challenges related to computational complexity, scalability, and system reliability. This paper proposes an optimization model for the Borda Count method by integrating blockchain consensus mechanisms, including Proof of Work (PoW), Proof of Stake (PoS), and Byzantine Fault Tolerance (BFT), aiming to enhance the voting process's efficiency, accuracy, and fault tolerance.We explore how blockchain technology can address the computational challenges of Borda Count, ensuring secure, transparent, and decentralized voting while maintaining high system reliability. By leveraging blockchain's immutability and consensus mechanisms, the proposed model significantly reduces computational overhead, increases the robustness of the system against node failures, and improves the accuracy of the voting results.This paper presents an in-depth analysis of the Borda Count method and blockchain consensus mechanisms, outlines a novel optimization algorithm, and provides a theoretical evaluation of the model's performance. We conclude by discussing the advantages of integrating blockchain with Borda Count for distributed voting systems and suggest potential directions for future research.
Stefan A. Iordache, Catalina Camelia Patilea, Ciprian Păduraru
In recent years, the rapid development of autonomous vehicles (AVs) has brought new challenges in terms of data security, privacy, and communication integrity. Our research investigates the potential of blockchain technology to improve the security of AVs by securing vehicle communication systems. By integrating blockchain with AI-based predictive algorithms, this approach aims to secure vehicle peer-to-peer communication, reduce traffic congestion, and improve safety for drivers and pedestrians. Blockchain’s decentralized ledger ensures the integrity of data exchange between vehicles and smart city infrastructure and mitigates the risks of cyberattacks such as data manipulation and identity forgery. This paper also examines recent advances in vehicular ad hoc networks (VANETs) and vehicular social networks (VSNs), and it demonstrates how the immutability and cryptographic security of the blockchain can strengthen AV systems. The proposed architecture not only protects user privacy but also decentralizes access to critical data needed for AI-driven decisions, ultimately promoting a safer and more reliable environment for autonomous vehicles.
Yalan Wang, Liqun Chen, Long Meng, Christopher J. P. Newton
Concerns about how third parties manage personal information have led to the development of decentralized identities (DIDs) and verifiable credentials (VCs). The World Wide Web Consortium (W3C) working group has been developing standards for DIDs and VCs. In the W3C standards, a DID identifies an entity (a DID holder) and a VC confirms that this DID holder has some associated attributes. A DID holder can obtain many VCs and confirm any number of these VCs to others (verifiers) in verifiable presentations (VPs). In order to keep a holder’s identity and attributes private, it is necessary to achieve anonymous VPs that allows this information to be kept confidential. The W3C working group recommends using randomizable signatures to create VCs with zero-knowledge proofs for this purpose. However, the anonymous VPs provided by the this method are limited that in the real world, credentials in cross domains cannot be universally verified. To overcome this limitation, in this paper, we propose a new scheme, called Verifiable Credentials with anonymous DIDs (VCaDID), which aims to achieve anonymous VPs in cross-domain settings. The main technique in our VCaDID scheme is a ring signature with multiple attributes by hiding a holder’s public key among a ring of holders. In our scheme, we set private keys associated with the holder’s DID and attributes, which allow the holder to anonymously present these credentials in a verifiable way. We also prove that the proposed VCaDID scheme satisfies correctness, anonymity and unforgeability under security assumptions of discrete log and random oracle model. Finally, we implement our scheme to demonstrate its feasibility.
With the rapid advancement and application of the Internet of Medical Things (IoMT), personal health records (PHRs) are now increasingly comprised of data collected by Internet of Things (IoT) devices and medical records documented by healthcare professionals. Personal health record (PHR) sharing demonstrates great potential in improving the accuracy of disease diagnosis. However, PHR sharing also brings risks such as illegal access and personal information leakage. Some works explored using blockchain or attribute-based encryption (ABE) to solve these privacy leakage problems, but those solutions did not pay attention to the user’s attribute privacy. In this work, we combine a linear secret sharing scheme (LSSS) and zero-knowledge succinct non-interactive argument of knowledge (zkSNARK) scheme to design an efficient zero-knowledge proof protocol called zk-AHSNARK. It can verify the user’s attribute permissions while also hiding attribute information. Based on zk-AHSNARK, we propose a novel PHR sharing scheme that protects attribute privacy. Data security is ensured by storing encrypted data in the interplanetary file system (IPFS). In addition, we introduce keyword ciphertext search to achieve fast data retrieval, and we implement the search and verification algorithms via a smart contract, ensuring the trustworthiness and integrity of the execution. Finally, through a large number of simulations, we demonstrated the suggested scheme’s viability and security.
Decentralized applications (DApps) are increasingly using off-chain data, yet growing concerns about data privacy hinder their widespread adoption. Zero-knowledge proofs (ZKPs) have emerged as a solution to this problem. This paper proposes a novel hybrid framework that combines Zero-Knowledge Scalable Transparent Arguments of Knowledge (zk-STARKs) and Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge (zk-SNARKs) to deliver an efficient, scalable, quantum-resistant, and privacy-preserving ZKP system. The framework employs zk-STARKs' ability to handle large computations with quantum resistance and zk-SNARKs' succinct proofs and fast verification. The hybrid framework utilizes advanced arithmetization techniques to balance scalability, privacy, and security, including Algebraic Intermediate Representation (AIR) for zk-STARKs and Rank-1 Constraint Systems (RlCS) or PLONKish constraints for zk-SNARKs. The framework uses the Kate-Zaverucha-Goldberg (KZG) polynomial commitment scheme for reliability and transparency and eliminates trusted setup by using Discrete-logarithm-based Argument of Recursive Knowledge (DARK) commitments and Poseidon hashing. This paper details the construction of the hybrid proof system, analyzes its complexity, and explores its potential applications. The framework solves smart contract data feed limitations, security, efficiency, and scalability while prioritizing privacy.
With the development of vehicular networks and cloud-edge collaborative technologies, a large amount of vehicle data is collected at edge nodes (Edge Node, EN) for analysis and decision-making. However, edge data faces challenges in terms of integrity and security. Data owners (Data Owner, DO) should delegate auditors to periodically verify the integrity of the data. However, existing verification methods have not yet addressed issues related to verifiers forging evidence and fair payment. This paper proposes a smart contract-based edge data integrity verification scheme. An audit tree based on lattice hashing is designed, allowing the smart contract to initiate multiple verification challenges while only storing a complete label, thus reducing storage overhead. The homomorphic additivity of lattice hashing supports arbitrary data segmentation as challenges, effectively preventing EN from forging evidence. This scheme also designs two smart contract arbitration algorithms to ensure fair payment. Experimental comparisons show that this scheme effectively resolves the trust issues related to EN and ensures fair payment among EN, CSP, and DO.
Ze-Peng Yang, Wenlong Tian, Emma Zhang, Zhiyong Xu
With the widespread adoption of cloud storage, effectively identifying and eliminating redundant data among users while ensuring data security has become a significant challenge. However, traditional similarity detection methods has limitations in privacy protection. Although conventional encryption techniques can safeguard privacy, they have difficulty detecting redundancy between similar blocks. Thus, we propose a secure reduction of redundant and similar data for cloud storage to address these challenges based on zero-knowledge proof (Sec-Reduce), called Sec-Reduce. It first employs a novel zero-knowledge proof technique for file-level redundancy detection, where redundant files are identified and excluded from storage. To further determine the similarity of non-redundant files, the scheme performs content-based chunking and feature extraction using a similarity feature extraction method. These extracted features are then encrypted using the approximate homomorphic encryption scheme Cheon-Kim-Kim-Song (CKKS) to enable similarity detection in the ciphertext environment. Finally, secure delta encoding is applied to store unique ciphertext blocks and deltas. Evaluations of real-world datasets demonstrate that Sec-Reduce achieves higher storage savings than existing encrypted storage methods, with storage overhead comparable to plaintext storage and only moderate performance overhead.
Privacy-Preserving Coalesced Learning (PPCL) exist a distributed method to machine learning that allows various plans or administrations to collaboratively order models deficiency allocation raw source, shielding separable privacy. Unlike old-style centralized methodologies, PPCL uses methods like protected collections, other concealment, and homomorphic encryption en route for stopping subtle source leakage. Secure aggregation syndicates separate idea updates securely, discrepancy privacy vaccinates noise to avoid model re-identification, and homomorphic encryption allows encrypted calculations. Despite its assurance, PPCL faces several tests. Communication above from recurrent informs can be heavy, scalability problems arise as contributor number grows, and combative outbreaks may deed system susceptibilities. Additionally, device and data heterogeneity present hurdles to attaining steady truthfulness and impartiality. Future research in PPCL aims to address these tests by improving scalability, enhancing communication, and improving security. These progressions could enlarge PPCL's application to privacy-sensitive areas like healthcare and finance, supporting secure, decentralized data-driven inventions.
Antonio Jesús Chaves, Cristian Martín, Kwang Soon Kim, Adnan Shahid · 5 authors
Machine learning data privacy has been improved with Federated Learning approaches. However, some obstacles to guaranteeing traceability, openness, and participant contribution incentives prevent its widespread use. In this study, Ethereum blockchain technology is integrated into the data stream Kafka-ML framework, presenting a novel asynchronous and blockchain-based Federated Learning approach. By utilising Ethereum for transparent and auditable participant tracking, this integration overcomes some shortcomings such as auditability and model sharing reliability. Furthermore, Ethereum smart contracts allow for automatic reward distribution systems, which promote equitable incentive systems and increased involvement in the Federated Learning process. To demonstrate its potential, an extensive evaluation has been carried out on a wireless net-work technology detection use case. By improving transparency, traceability, and incentive structures of Federated Learning, it is expected to strengthen the robustness of flexible machine learning collaboration with data streams.
Federated Learning (FL) enables decentralized model training while ensuring data privacy between clients and the server, and is applicable in various domains such as healthcare, finance, and edge computing. Traditional FL approaches assume homogeneous model architectures across clients and servers, limiting their applicability in real-world scenarios where clients use diverse models. To address this, we introduce Feature Align-HFL, a novel framework designed for Heterogeneous Federated Learning (HFL) that aligns dissimilar models for effective knowledge sharing. Specifically, we demonstrate its application by aligning CNNs and U-Nets for a common classification task. By employing cosine similarity, Feature Align-HFL identifies and aligns common features across heterogeneous models. Experiments on the CIFAR-10 dataset show that CNN and adapted U-Net models achieved 90 percent and 86 percent accuracy, respectively, with feature map similarities reaching up to 96 percent. These results highlight the significant transferable knowledge between dissimilar architectures, indicating that Feature Align-HFL can effectively support knowledge sharing in heterogeneous FL settings.
Dongliang Cai, Liang Zhang, Borui Chen, Haibin Kan
Decentralized data sovereignty and secure data exchange are regarded as foundational pillars of the new era. Attribute-based encryption (ABE) is a promising solution that enables fine-grained access control in data sharing. Recently, Hohenberger et al. (Eurocrypt 2023) introduced registered ABE (RABE) to eliminate trusted authority and gain decentralization. Users generate their own public and secret keys and then register their keys and attributes with a transparent key curator. However, RABE still suffers from heavy decryption overhead. A natural approach to address this issue is to outsource decryption to a decryption cloud server (DCS). In this work, we propose the first auditable RABE scheme with reliable outsourced decryption (ORABE) based on blockchain. First, we achieve verifiability of transform ciphertext via a verifiable tag mechanism. Then, the exemptibility, which ensures that the DCS escapes false accusations, is guaranteed by zero knowledge fraud proof under the optimistic assumption. Additionally, our system achieves fairness and auditability to protect the interests of all parties through blockchain. Finally, we give concrete security and theoretical analysis and evaluate our scheme on Ethereum to demonstrate feasibility and efficiency.
To achieve fine-grained access control and address the data silos challenge in data sharing, the integration of blockchain with attribute-based encryption emerges as a promising solution. Nowadays, the growing interconnectedness among diverse blockchain applications has spurred the need for efficient cross-chain data sharing. However, existing single-authority attribute-based data sharing schemes are not suitable for such cross-chain scenarios involving multiple attribute authorities. Moreover, the frequent requirement for data owners to process cross-chain data requests significantly hampers practicality. In this context, we introduce a novel multi-authority attribute-based proxy re-encryption scheme that enables ciphertext policy updating and supports secure and efficient cross-chain data sharing. By introducing a proxy, the data owner is empowered to delegate access without leaking any valid information and flexibly sells data across blockchains through cross-chain access policies. Besides, our scheme leverages the relay chain to foster a decentralized and trustworthy ecosystem. The adoption of smart contracts automates the cross-chain data sharing process and ensures equitable distribution of benefits among participants. Additionally, our scheme integrates hybrid encryption with the decentralized data hosting platform, substantially mitigating the on-chain storage burden. Security analysis affirms that our scheme is semantically secure and resistant to collusion attack. Performance analysis and simulation experiments demonstrate the excellent efficiency and practicality of our scheme when conducting cross-chain data sharing.
Intelligent Transport Systems are designed to revolutionize the performance and efficiency of Vehicular Ad Hoc Networks (VANETs). Consumer electronics encompassing autonomous driving and route planning contribute to a better driving experience. Federated learning enables vehicle collaboration to train global models of intelligent transport without sharing local data for personalized consumer electronics. However, due to the dynamic network topology and unreliable open channels of VANETs, various potential risks undermine the credibility of establishing intermediate model parameters. To address these issues, we propose a lightweight authentication and privacy-preserving aggregation scheme for blockchain-enabled federated learning in VANETs(LPBFL). Specifically, we first construct a distributed secure authentication framework for blockchain federated learning that facilitates seamless authentication of mobile vehicles and security of model transmission. Subsequently, we design a lightweight three-party authentication key agreement based on chaotic map that establishes a session key for secure transmission of the intermediate model. In addition, we propose a continuous authentication adaptive model aggregation algorithms to ensure local model integrity while improving the quality of global models. Finally, security analysis and proofs show that LPBFL enhances the privacy and reliability of intermediate model parameters. Comprehensive experimental evaluations substantiate that the proposed LPBFL facilitates lightweight authentication while maintaining superior model accuracy.