This paper aims to investigate the convergence of Federated Learning (FL) and Blockchain technology to enhance the security and robustness of machine learning systems. As the utilization of mobile devices grows, managing the complexities of unbalanced and non-independent data becomes imperative. Fed-erated Learning introduces a decentralized training approach, while Blockchain provides transparency and immutability. This study proposes a resilient solution for safeguarding FL against single points of failure (SPoF) and poisoning attacks, integrating a decentralized validation mechanism and a Proof-of-Stake consen-sus protocol. Extensive experimentation is conducted to assess the effectiveness of the proposed framework across varying degrees of adversarial device engagement. The results demonstrate the solution's efficacy in fortifying FL systems against malicious participants, with an impact observed up to 50% adversarial involvement. This underscores the potential of the approach to elevate security in the realm of decentralized machine learning.
This research paper delves into the imperative domain of bolstering data confidentiality within smart contracts through the integration of advanced privacy-preserving methodologies. Smart contracts, pivotal components of blockchain technology, execute self-executing contracts with predefined conditions and are increasingly utilized across various sectors, necessitating stringent data protection measures. The paper addresses the pressing need for fortified data privacy within smart contracts and investigates cutting-edge approaches to mitigate privacy challenges. Two focal techniques under scrutiny are zero-knowledge proofs (SBÇs) and homomorphic encryption. SBÇs facilitate the validation of computations without revealing sensitive data, enabling parties to verify transaction authenticity without disclosing the underlying information. Meanwhile, homomorphic encryption permits computations on encrypted data, preserving confidentiality by allowing operations on encrypted information without the need for decryption. By analyzing these advanced privacy-preserving techniques, this study aims to address the vulnerabilities in data confidentiality present in smart contracts. Its findings hold significant promise in fortifying the security and confidentiality of transactions, thus contributing substantially to the evolution of secure blockchain technology. This research underscores the pivotal role of innovative privacy-enhancing mechanisms in safeguarding sensitive data within smart contracts, ensuring the trust and integrity essential for their widespread adoption.
The fundamental theorem of Goldreich, Micali, and Wigderson (J. ACM 1991) shows that the existence of a one-way function is sufficient for constructing computational zero knowledge (CZK) proofs for all languages in NP. We prove its converse, thereby establishing characterizations of one-way functions based on the worst-case complexities of zero knowledge. Specifically, we prove that the following are equivalent: - A one-way function exists. - NP ⊆ CZK and NP is hard in the worst case. - CZK is hard in the worst case and the problem GapMCSP of approximating circuit complexity is in CZK. The characterization above also holds for statistical and computational zero-knowledge argument systems. We further extend this characterization to a proof system with knowledge complexity O(logn). In particular, we show that the existence of a one-way function is characterized by the worst-case hardness of CZK if GapMCSP has a proof system with knowledge complexity O(logn). We complement this result by showing that NP admits an interactive proof system with knowledge complexity ω(logn) under the existence of an exponentially hard auxiliary-input one-way function (which is a weaker primitive than an exponentially hard one-way function). We also characterize the existence of a robustly-often nonuniformly computable one-way function by the nondeterministic hardness of CZK under the weak assumption that PSPACE ⊈AM. We present two applications of our results. First, we simplify the proof of the recent characterization of a one-way function by NP-hardness of a meta-computational problem and the worst-case hardness of NP given by Hirahara (STOC’23). Second, we show that if NP has a laconic zero-knowledge argument system, then there exists a public-key encryption scheme whose security can be based on the worst-case hardness of NP. This improves previous results which assume the existence of an indistinguishable obfuscation.
Nir Bitansky, Chethan Kamath, Omer Paneth, Ron D. Rothblum · 5 authors
Batch proofs are proof systems that convince a verifier that x1,…,xt ∈ L, for some NP language L, with communication that is much shorter than sending the t witnesses. In the case of statistical soundness (where the cheating prover is unbounded but the honest prover is efficient given the witnesses), interactive batch proofs are known for UP, the class of unique-witness NP languages. In the case of computational soundness (where both honest and dishonest provers are efficient), non-interactive solutions are now known for all of NP, assuming standard lattice or group assumptions. We exhibit the first negative results regarding the existence of batch proofs and arguments: - Statistically sound batch proofs for L imply that L has a statistically witness indistinguishable (SWI) proof, with inverse polynomial SWI error, and a non-uniform honest prover. The implication is unconditional for obtaining honest-verifier SWI or for obtaining full-fledged SWI from public-coin protocols, whereas for private-coin protocols full-fledged SWI is obtained assuming one-way functions. This poses a barrier for achieving batch proofs beyond UP (where witness indistinguishability is trivial). In particular, assuming that NP does not have SWI proofs, batch proofs for all of NP do not exist. - Computationally sound batch proofs (a.k.a batch arguments or BARGs) for NP, together with one-way functions, imply statistical zero-knowledge (SZK) arguments for NP with roughly the same number of rounds, an inverse polynomial zero-knowledge error, and non-uniform honest prover. Thus, constant-round interactive BARGs from one-way functions would yield constant-round SZK arguments from one-way functions. This would be surprising as SZK arguments are currently only known assuming constant-round statistically-hiding commitments. We further prove new positive implications of non-interactive batch arguments to non-interactive zero knowledge arguments (with explicit uniform prover and verifier): - Non-interactive BARGs for NP, together with one-way functions, imply non-interactive computational zero-knowledge arguments for NP. Assuming also dual-mode commitments, the zero knowledge can be made statistical. Both our negative and positive results stem from a new framework showing how to transform a batch protocol for a language L into an SWI protocol for L.
We put forward a new approach for achieving non-interactive zero-knowledge proofs (NIKZs) from the learning with errors (LWE) assumption (with subexponential modulus to noise ratio). We provide a LWE-based construction of a hidden bits generator that gives rise to a NIZK via the celebrated hidden bits paradigm. A notable feature of our construction is its simplicity. Our construction employs lattice trapdoors, but beyond that uses only simple operations. Unlike prior solutions, we do not rely on a correlation intractability argument nor do we utilize fully homomorphic encryption techniques. Our solution provides a new methodology that adds to the diversity of techniques for solving this fundamental problem.
Identity management plays a critical role in Web3 applications. Decentralized Identity (DID) offers a privacy-preserving solution, giving users full control over their identity information. Existing research on DID primarily focuses on single-owner scenarios, where owners have complete privileges for owner management and credentials. However, in multi-owner cases, current coarse-grained identity management approaches lead to serious privacy and security problems, such as identity impersonation and high key recovery overhead. Little work has been done on identity management for multiple owners. In this paper, we propose MoDID, a fine-grained identity management scheme for multiple owners, which complies with the DID standard proposed by W3C. First, our solution allows multiple owners to control DID subjects flexibly and reliably through hierarchical owner management. Additionally, we design a secure key recovery scheme to reduce the risk of identity loss while introducing lower overhead. Finally, we implement MoDID on the Sepolia Ethereum Test Network to evaluate the effectiveness of our proposed scheme. The result demonstrates that our system allows multiple owners to manage a single identity with lower gas consumption and time consumption than the state-of-the-art.
This study examines the issues of privacy protection, data security, and query efficiency in blockchain-based electronic medical record (EMR) sharing. It proposes a secure storage and sharing scheme for EMR based on Hyperledger Fabric and the InterPlanetary File System (IPFS). To mitigate the privacy risks of data mining that could reveal patient identities, we establish an attribution channel in Hyperledger Fabric to store EMR ownership information and a data channel to store the storage location, digest, and usage records of medical data. Encrypted medical data are stored in the IPFS. To improve query efficiency in the blockchain, we integrate queryable medical data attributes into a composite key for conditional queries, avoiding complex data filtering processes. Additionally, we use a zero-knowledge proof combined with smart contracts for decentralized identity verification, eliminating reliance on third-party centralized verification services and enhancing system security. We also integrate AES and proxy re-encryption techniques to ensure data security during sharing. This scheme provides a more secure, efficient, and privacy-preserving approach for EMR systems, with significant practical implications and broad application potential.
William Boitier, Antonella Del Pozzo, Álvaro García-Pérez, Stéphane Gazut · 12 authors
Federated Learning is a decentralized framework that enables multiple clients to collaboratively train a machine learning model under the orchestration of a central server without sharing their local data. The centrality of this framework represents a point of failure which is addressed in literature by blockchain-based federated learning approaches. While ensuring a fully-decentralized solution with traceability, such approaches still face several challenges about integrity, confidentiality and scalability to be practically deployed. In this paper, we propose Fantastyc, a solution designed to address these challenges that have been never met together in the state of the art.
Kulvinder Singh, R. P. S. Tomar, Vansh Chaudhary, Mridul Jain
The cloud-based meteorological data management offers vast potential for research and collaboration, but inherent privacy concerns and data integrity risks necessitate robust solutions. This research suggests a new, multi-layered method to improve integrity and privacy in this field. By utilizing homomorphic encryption, this study makes it possible to perform calculations on encrypted meteorological data while keeping private information hidden from cloud service providers. Federated learning promotes cooperation without compromising the privacy of raw data by enabling distributed learning on local datasets. Lastly, zero-knowledge proofs are integrated to confirm the accuracy of computations made on encrypted data, ensuring the validity of outcomes without disclosing the underlying data. This integrated strategy presents a viable path for reliable and safe cloud-based meteorological data administration, promoting cutting-edge study and creativity while protecting private data.
The land registration procedures in India face challenges due to fragmented document maintenance across departments, causing delays and errors in verifying land ownership. This study introduces a solution, “Secured Land Registration using Ethereum Blockchain and IPFS,” tailored for India. This study utilizes advanced technologies such as Ethereum blockchain, IPFS, smart contracts, Solidity, and the Sepolia TestNet to address existing system limitations. Integration with Ethereum blockchain enhances security and trust through its decentralized and tamper-resistant features. Smart contracts, coded in Solidity, automate processes, introducing efficiency gains. The Sepolia TestNet acts as a testing ground, ensuring robustness before deployment on the Ethereum network. A significant innovation involves integration with IPFS using Pinata API for decentralized and immutable document storage, addressing challenges related to data integrity and accessibility. This project envisions a future where land registration in India becomes secure, accessible, and efficient. The combination of blockchain, IPFS, and smart contracts provides a user-centric, streamlined, and technologically advanced approach, overcoming traditional method limitations. This research study details the methodology for system implementation, offers a thorough comparison with traditional land registration systems, and presents results, providing insights into the evolution of land registration processes in the Indian real estate landscape.
Medical healthcare centers are envisioned as a promising paradigm to handle vast data for various disease diagnoses using artificial intelligence. Traditional Machine Learning algorithms have been used for years, putting the sensitivity of patients' medical data privacy at risk. Collaborative data training, where multiple hospitals (nodes) train and share encrypted federated models, solves the issue of data leakage and unites resources of small and large hospitals from distant areas. This study introduces an innovative framework that leverages blockchain-based Federated Learning to identify 15 distinct lung diseases, ensuring the preservation of privacy and security. The proposed model has been trained on the NIH Chest Ray dataset (112 120 X-Ray images), tested, and evaluated, achieving test accuracy of 92.86%, a latency of 43.518625 ms, and a throughput of 10034017 bytes/s. Furthermore, we expose our framework blockchain to stringent empirical tests against leading cyber threats to evaluate its robustness. With resilience metrics consistently nearing 87% against three evaluated cyberattacks, the proposed framework demonstrates significant robustness and potential for healthcare applications. To the best of our knowledge, this is the first paper on the practical implementation of blockchain-empowered FL with such data and several diseases, including multiple disease coexistence detection.
Blockchain Technology Applications and Security
Privacy-Preserving Technologies in Data
Artificial Intelligence in Healthcare and Education
In order to achieve zero-knowledge proof (ZKP) in distributed mobile scenarios, we propose a two-stage multi-prover ZKP framework. Our method utilizes secure multi-party computation (MPC), which has advantages such as flexible adaptation, stable performance, and fewer restrictions compared to existing solutions. In addition, based on the properties of cyclic groups, we optimize secure multi-party summation, improving the balance between security and efficiency, as well as transferability of the algorithm.
The rapid growth of increasingly pervasive smart devices with smart applications has contributed to numerous wireless security vulnerabilities in consumer electronics, which compromises the integrity of the entire network. Attacks frequently involve unauthorized access to inadequate wireless Internet of Things(IoT) based consumer electronic devices, information security lapses, and hence leakage of private and sensitive data occurs. Due to the centralized architecture of IoT networks, data security depends on the owners’ trust in the entities. With the potential advantages of blockchain technology, a distributed ledger eliminates the limitations of centralized architecture while establishing trust, privacy, and security with assured reliability. This paper proposes a smart contract-based decentralized Selective Ring-Role-based access control framework for real-time health monitoring in consumer electronic networks. The proposed framework incorporates off-chain AI-driven anomaly detection for reduced network load and latency. The experimental results show that the proposed smart contract framework is robust in proving the security and privacy of IoT-based consumer electronic networks with improved anomaly detection over the state-of-the-art methods.
Federated learning allows several clients to train one machine learning model jointly without sharing private data, providing privacy protection. However, traditional federated learning is vulnerable to poisoning attacks, which can not only decrease the model performance, but also implant malicious backdoors. In addition, direct submission of local model parameters can also lead to the privacy leakage of the training dataset. In this paper, we aim to build a privacy-preserving and Byzantine-robust federated learning scheme to provide an environment with no vandalism (NoV) against attacks from malicious participants. Specifically, we construct a model filter for poisoned local models, protecting the global model from data and model poisoning attacks. This model filter combines zero-knowledge proofs to provide further privacy protection. Then, we adopt secret sharing to provide verifiable secure aggregation, removing malicious clients that disrupting the aggregation process. Our formal analysis proves that NoV can protect data privacy and weed out Byzantine attackers. Our experiments illustrate that NoV can effectively address data and model poisoning attacks, including PGD, and outperforms other related schemes.
Tahani Baabdullah, Amani Alzahrani, Danda B. Rawat, Chunmei Liu
Increasing global credit card usage has elevated it to a preferred payment method for daily transactions, underscoring its significance in global financial cybersecurity. This paper introduces a credit card fraud detection (CCFD) system that integrates federated learning (FL) with blockchain technology. The experiment employs FL to establish a global learning model on the cloud server, which transmits initial parameters to individual local learning models on fog nodes. With three banks (fog nodes) involved, each bank trains its learning model locally, ensuring data privacy, and subsequently sends back updated parameters to the global learning model. Through the integration of FL and blockchain, our system ensures privacy preservation and data protection. We utilize three machine learning and deep neural network learning algorithms, RF, CNN, and LSTM, alongside deep optimization techniques such as ADAM, SGD, and MSGD. The SMOTE oversampling technique is also employed to balance the dataset before model training. Our proposed framework has demonstrated efficiency and effectiveness in enhancing classification performance and prediction accuracy.
Today, blockchain ledgers utilize concurrent deterministic execution schemes to scale up. However, ordering fairness is not preserved in these schemes: although they ensure all replicas achieve the same serial order, this order does not always align with the fair, consensus-established order when executing smart contracts with runtime-determined accesses. To preserve ordering fairness, an intuitive method is to concurrently execute transactions and re-execute any order-violating ones. This in turn increases unforeseen conflicts, leading to scaling bottlenecks caused by numerous costly aborts under contention. To address these issues, we propose Spectrum, a novel deterministic execution scheme for smart contract execution on blockchain ledgers. Spectrum preserves the consensus-established serial order (so-called strict determinism) with high performance. Specifically, we leverage a speculative deterministic concurrency control to execute transactions in speculation and enforce an agreed-upon serial order by aborting and re-executing any mis-speculated ones. To overcome the scaling bottleneck, we present two key optimizations based on speculative processing: operation-level rollback and predictive scheduling, for reducing both the overhead and the number of mis-speculations. We evaluate Spectrum by executing EVM-based smart contracts on popular benchmarks, showing that it realizes fair smart contract execution by preserving ordering fairness and outperforms competitive schemes in contended workloads by 1.4x to 4.1x.
Huong Q. Nguyen, Tri Nguyen, Lauri Lovén, Susanna Pirttikangas
This paper presents a fully coupled blockchain-assisted federated learning architecture that effectively eliminates single points of failure by decentralizing both the training and aggregation tasks across all participants. Our proposed system offers a high degree of flexibility, allowing participants to select shared models and customize the aggregation for local needs, thereby optimizing system performance, including accurate inference results. Notably, the integration of blockchain technology in our work is to promote a trustless environment, ensuring transparency and non-repudiation among participants when abnormalities are detected. To validate the effectiveness, we conducted real-world federated learning deployments on a private Ethereum platform, using two different models, ranging from simple to complex neural networks. The experimental results indicate comparable inference accuracy between centralized and decentralized federated learning settings. Furthermore, our findings indicate that asynchronous aggregation is a feasible option for simple learning models. However, complex learning models require greater training model involvement in the aggregation to achieve high model quality, instead of asynchronous aggregation. With the implementation of asynchronous aggregation and the flexibility to select models, participants anticipate decreased aggregation time in each communication round, while experiencing minimal accuracy trade-off.
Venkata Naga Rani Bandaru, P. Visalakshi, L. N. Prakash Kumar Ponnuru, Shaik Mohammad Rafee · 5 authors
The synergy between machine learning and encryption fortifies data security and privacy. This comprehensive overview delves into pivotal encryption methods in ML, spotlighting their inherent adaptability and paramount role in shielding sensitive data. Differential privacy injects controlled noise, ensuring privacy preservation while upholding data utility, especially vital in healthcare and financial sectors. Federated learning facilitates decentralized training, while homomorphic encryption assures secure data processing. Secure multi-party computation (SMPC) empowers collaborative private computation, and zero-knowledge proofs authenticate veracity sans data exposure, pivotal in blockchain and identity validation. These sophisticated algorithms cater comprehensively to diverse security requisites, bolstering data protection across indispensable domains.
Anuj Nepal, Mohamed Ahzam Amanullah, Robin Doss, Frank Jiang
The evolution of the Internet of Vehicles (IoVs) presents many opportunities for intelligent transport systems; however, it brings significant security challenges that threaten the security and reliability of the data. Security, reliability, and trustworthiness are the essential critical requirements for the IoVs to ensure secure and efficient decision-making processes in an accurate, secure, and trustworthy manner. Traditional research paradigms have predominantly focused on data integrity, overlooking the essential need for data to be realistic, consistent, trustworthy, and reliable. Addressing this gap, our paper introduces a decentralized secure data provenance (SDP) protocol for the dynamic and distributed nature of IoVs to ensure verifiable security properties regarding data plausibility, source identity, data privacy, location authenticity, and data integrity, properties that are fundamental for achieving SDP. Our protocol integrates Verifiable Credentials (VCs) with distributed ledger technology (DLT), Road-Side Unit (RSU) infrastructure, cryptographic techniques, and plausibility checks to ensure secure, traceable, and tamper-evident data lineage. This comprehensive approach enables the timely identification and mitigation of potential security breaches, such as impersonation, data tampering, and privacy violations which are crucial for maintaining SDP and ensuring data credibility through anomaly detection and plausibility checks. Through detailed security analysis and validation, our methodology demonstrates improved resilience against common threats, ensuring that only credible and verifiable data inform its decision-making processes.