This paper presents an in-depth examination of privacy-enhancing methodologies in machine learning. It highlights the integration of federated learning with cutting-edge encryption techniques and explores how blockchain architectures contribute to data privacy. A major focus is on federated learning, a decentralized model training strategy, and its combination with privacy-protecting technologies like Homomorphic Encryption, Differential Privacy, and Secure Multi-Party Computation. We emphasize that federated learning naturally improves data privacy and, when paired with cryptographic methods, increases resilience against data breaches and cyber-attacks. Additionally, this study explores the potential of blockchain in enhancing data privacy. Blockchain's immutable and transparent characteristics, supplemented with shuffling technology, zero-knowledge proofs, and ring signatures, improve the confidentiality and integrity of data transactions. The paper also emphasizes the critical need for transparency and explainability in machine learning, advocating for methods that demystify the decision-making processes of ML models. This transparency is crucial for building trust and is becoming a regulatory requirement in many industries. Furthermore, the paper discusses the importance of auditing in machine learning, highlighting the need for comprehensive model validation and ethical considerations. In conclusion, the paper argues that achieving a balance 1 between functionality and privacy in ML applications is essential. It suggests that a combination of federated learning, advanced cryptographic techniques, and explainable AI principles can create effective and privacy-respecting systems.
Blockchain technology has been incorporated into the Healthcare Internet of Things (IoT) landscape as a revolutionary solution to tackle issues related to the sharing of medical records. This paper presents an innovative method that utilizes Temporal Blockchain for the purpose of Provenance Tracking. The introductory section provides context by delineating the significance of trust and transparency in medical data sharing within the healthcare IoT ecosystem. The study examines current blockchain solutions, delving into frameworks such as Hyperledger Fabric, Ethereum, Corda, and specialized approaches like temporal blockchain. The paper examines the difficulties associated with tracking the origin of data, concerns regarding privacy, problems related to scalability, and the need to comply with regulations. These challenges provide the context for the proposed methodology. The main emphasis is on Temporal Blockchain, integrating temporal elements to improve the tracking of origin and history. The evaluation parameters, such as security, provenance tracking, scalability, interoperability, privacy compliance, and performance, undergo a thorough assessment. The attained values demonstrate a strong emphasis on security at a high level, thorough tracking of origin and history, and strict adherence to privacy regulations. Nevertheless, the need for scalability and interoperability necessitates meticulous consideration. The study showcases the capacity of Temporal Blockchain to establish trust and enhance transparency in the sharing of medical records. The future scope focuses on tackling scalability challenges, improving interoperability, and making continuous optimization efforts. The proposed approach highlights notable accomplishments and emphasizes the continuous development and collaborative aspect of Blockchain-Based Medical Record Sharing in Healthcare IoT.
Open access
Blockchain Technology Applications and Security
Privacy-Preserving Technologies in Data
Artificial Intelligence in Healthcare and Education
Abstract The article aims to investigate the potential of blockchain technology in mitigating certain cybersecurity risks associated with artificial intelligence (AI) systems. Aligned with ongoing regulatory deliberations within the European Union (EU) and the escalating demand for more resilient cybersecurity measures within the realm of AI, our analysis focuses on specific requirements outlined in the proposed AI Act. We argue that by leveraging blockchain technology, AI systems can align with some of the requirements in the AI Act, specifically relating to data governance, record-keeping, transparency and access control. The study shows how blockchain can successfully address certain attack vectors related to AI systems, such as data poisoning in trained AI models and data sets. Likewise, the article explores how specific parameters can be incorporated to restrict access to critical AI systems, with private keys enforcing these conditions through tamper-proof infrastructure. Additionally, the article analyses how blockchain can facilitate independent audits and verification of AI system behaviour. Overall, this article sheds light on the potential of blockchain technology in fortifying high-risk AI systems against cyber risks, contributing to the advancement of secure and trustworthy AI deployments. By providing an interdisciplinary perspective of cybersecurity in the AI domain, we aim to bridge the gap that exists between legal and technical research, supporting policy makers in their regulatory decisions concerning AI cyber risk management.
This paper presents a cryptographic solution for establishing trust in peer-to-peer (P2P) networks, addressing issues of privacy, performance, and anonymity. Our protocol utilizes Zero-Knowledge Proofs (ZKP) for continuous trust validation during data transfers. This procedure compels each node to continually demonstrate its integrity, significantly decreasing the potential for network at- tacks. Upon evaluation, the protocol proved to be highly scalable and efficient, expanding network reach without requiring additional control messages. This result validates the protocolâs robustness, suggesting its potential use in larger and more intricate P2P network architectures.
The technological advancements in the field of E-healthcare have resulted in unprecedented generation of medical data which increases the risk of data security and privacy. Ensuring the privacy of Electronic Health Records (EHR) has become challenging due to outsourcing of healthcare information in the cloud. This increases the chance of data leakage to unauthorized users and affects the privacy and integrity of the user data. It requires a trustworthy central authority to protect the sensitive patient information from both internal and external attacks. This paper presents a blockchain based privacy preservation framework for securing EHR data. The proposed framework integrates the immutability and decentralized nature of blockchain with advanced cryptographic techniques to ensure the confidentiality, integrity and availability of EHR. The EHR data are stored in an InterPlanetary File System (IPFS) which is encrypted using a hybrid cryptographic algorithm. In addition, a novel smart contact based patient-centric access control is designed in this paper using a blockchain-based SHA-256 hashing algorithm to protect the privacy of patient data. The experimental results show that the proposed framework enables secure sharing of health information between network users with improved data privacy and security. Furthermore, the optimized search process reduces the time and space complexity compared to the traditional search process. Through the utilization of smart contracts, this framework enforces patient-centric access controls and allows patients to manage and authorize access to their medical data.
Jan 19, 2024¡Proceedings of the 2024 Guangdong-Hong Kong-Macao Greater Bay Area International Conference on Digital Economy and Artificial Intelligence
With the widespread application of blockchain technology, various range proof protocols based on zero-knowledge proofs have been proposed. However, existing range proof protocols suffer from issues such as high communication overhead and computational complexity. Therefore, this paper introduces an efficient and secure Zero-Knowledge Set Membership Proof Protocol (ZSMPP) to address these challenges. Building upon improvements to the proof structure of range proof protocols, the paper integrates the SM2 identity-based digital signature algorithm, effectively avoiding the time-consuming bilinear pairing operations and reducing computational costs. The proposed protocol offers an efficient and secure solution for the given problem. Experimental results demonstrate that, compared to protocols proposed by Bootle, Deng, Mao, and others, the protocol presented in this paper exhibits superior computational efficiency, providing an efficient and secure solution for data security and individual privacy protection in the digital age.
Abstract The most important and difficult challenge the digital society has recently faced is ensuring data privacy and security in cloudâbased Internet of Things (IoT) technologies. As a result, many researchers believe that the blockchain's Distributed Ledger Technology (DLT) is a good choice for various clever applications. Nevertheless, it encountered constraints and difficulties with elevated computing expenses, temporal demands, operational intricacy, and diminished security. Therefore, the proposed work aims to develop a Decentralized Identifiable Distributed Ledger TechnologyâBlockchain (DIDLTâBC) framework that is intelligent and effective, requiring the least amount of computing complexity to ensure cloud IoT system safety. In this case, the Rabin algorithm produces the digital signature needed to start the transaction. The public and private keys are then created to verify the transactions. The block is then built using the DIDLT model, which includes the block header information, hash code, timestamp, nonce message, and transaction list. The primary purpose of the Blockchain Consent Algorithm (BCA) is to find solutions for numerous unreliable nodes with varying hash values. The novel contribution of this work is to incorporate the operations of Rabin digital data signature generation, DIDLTâbased blockchain construction, and BCA algorithms for ensuring overall data security in IoT networks. With proper digital signature generation, key generation, blockchain construction and validation operations, secured data storage and retrieval are enabled in the cloudâIoT systems. By using this integrated DIDLTâBCA model, the security performance of the proposed system is greatly improved with 98% security, less execution time of up to 150 ms, and reduced mining time of up to 0.98 s.
Andrea Flamini, Giada Sciarretta, Mario Scuro, Amir Sharif ¡ 6 authors
Verifiable credentials are a digital analogue of physical credentials. Their authenticity and integrity are protected by means of cryptographic techniques, and they can be presented to verifiers to reveal attributes or even predicates about the attributes included in the credential. One way to preserve privacy during presentation consists in selectively disclosing the attributes in a credential. In this paper we present the most widespread cryptographic mechanisms used to enable selective disclosure of attributes identifying two categories: the ones based on hiding commitments - e.g., mdl ISO/IEC 18013-5 - and the ones based on non-interactive zero-knowledge proofs - e.g., BBS signatures. We also include a description of the cryptographic primitives used to design such cryptographic mechanisms. We describe the design of the cryptographic mechanisms and compare them by performing an analysis on their standard maturity in terms of standardization, cryptographic agility and quantum safety, then we compare the features that they support with main focus on the unlinkability of presentations, the ability to create predicate proofs and support for threshold credential issuance. Finally we perform an experimental evaluation based on the Rust open source implementations that we have considered most relevant. In particular we evaluate the size of credentials and presentations built using different cryptographic mechanisms and the time needed to generate and verify them. We also highlight some trade-offs that must be considered in the instantiation of the cryptographic mechanisms.
Christian DelgadoâvonâEitzen, Luis Anido, Manuel J. FernĂĄndez Iglesias
The issuance and verification of academic certificates face significant challenges in the digital era. The proliferation of counterfeit credentials and the lack of a reliable, universally accepted system for issuing and validating them pose critical issues in the educational domain. Certificates, traditionally issued by centralized educational institutions using their proprietary systems, pose challenges for straightforward verification, generating uncertainty about the credibility of academic achievements. In addition to diplomas issued by academic entities, it is now necessary in virtually all professional fields to stay updated and obtain accreditation for certain skills or experiences, which is a determining factor in securing or enhancing employment. Yet, there is no platform available to consistently demonstrate these capabilities and experiences. This article introduces a novel model for issuing and verifying academic information using non-fungible tokens (NFTs) supported by blockchain technologies, focused on compliance with the General Data Protection Regulation (GDPR). It describes a model that grants control to the data subject, enabling the management of information access while adhering to key GDPR principles. Simultaneously, it remains compatible with existing systems within organizations, and is flexible in certifying various types of academic information. The implications of this model are discussed, emphasizing the importance of addressing privacy in blockchain-based applications.
Electronic tickets (e-tickets) are gradually being adopted as a substitute for paper-based tickets to bring convenience to customers, corporations, and governments. However, their adoption faces a number of practical challenges, such as flexibility, privacy, secure storage, and inability to deploy on IoT devices such as smartphones. These concerns motivate the current research on e-ticket systems, which seeks to ensure the unforgeability and authenticity of e-tickets while simultaneously protecting user privacy. Many existing schemes cannot fully satisfy all these requirements. To improve on the current state-of-the-art solutions, this paper constructs a blockchain-enhanced privacy-preserving e-ticket system for IoT devices, dubbed PriTKT, which is based on blockchain, structure-preserving signatures (SPS), unlinkable redactable signatures (URS), and zero-knowledge proofs (ZKP). It supports flexible policy-based ticket purchasing and ensures user unlinkability. According to the data minimization and revealing principle of GDPR, PriTKT empowers users to selectively disclose subsets of (necessary) attributes to sellers as long as the disclosed attributes satisfy ticket purchasing policies. In addition, benefiting from the decentralization and immutability of blockchain, effective detection and efficient tracing of double spending of e-tickets are supported in PriTKT. Considering the impracticality of existing e-tickets schemes with burdensome ZKPs, we replace them with URS/SPS or efficient ZKP to significantly improve the efficiency of ticket issuing and make it suitable for use on smartphones.
With the rapid development of artificial intelligence (AI) in the healthcare industry, the sharing of personal healthcare data plays an essential role in advancing medical AI. Unfortunately, personal healthcare data sharing is plagued by challenges like ambiguous data ownership and privacy leakage. Blockchain, which stores the hash of shared data on-chain and ciphertext off-chain, is treated as a promising approach to address the above issues. However, this approach lacks a flexible and reliable mechanism for incremental updates of the same case data. To avoid the overhead of authentication, access control, and rewards caused by on-chain data changes, we propose a blockchain and trusted execution environment (TEE)-based privacy-preserving sharing scheme for healthcare data that supports incremental updates. Based on chameleon hash and TEE, the scheme achieves reliable incremental updates and verification without changing the on-chain data. In the scheme, for privacy concerns, off-chain data are protected through symmetric encryption, whereas data verification, decryption, and computation are performed within TEE. The experimental results show the feasibility and effectiveness of the proposed scheme.
Realistic synthetic tabular data generation encounters significant challenges in preserving privacy, especially when dealing with sensitive information in domains like finance and healthcare. In this paper, we introduce \textit{Federated Tabular Diffusion} (FedTabDiff) for generating high-fidelity mixed-type tabular data without centralized access to the original tabular datasets. Leveraging the strengths of \textit{Denoising Diffusion Probabilistic Models} (DDPMs), our approach addresses the inherent complexities in tabular data, such as mixed attribute types and implicit relationships. More critically, FedTabDiff realizes a decentralized learning scheme that permits multiple entities to collaboratively train a generative model while respecting data privacy and locality. We extend DDPMs into the federated setting for tabular data generation, which includes a synchronous update scheme and weighted averaging for effective model aggregation. Experimental evaluations on real-world financial and medical datasets attest to the framework's capability to produce synthetic data that maintains high fidelity, utility, privacy, and coverage.
Marco Marcozzi, Ernestas Filatovas, Linas Stripinis, Remigijus PaulaviÄius
The consensus protocol plays a vital role in the performance and security of a specific Distributed Ledger Technology (DLT) solution. Currently, the traditional classification of consensus algorithms relies on subjective criteria, such as protocol families (Proof of Work, Proof of Stake, etc.) or other protocol features. However, such classifications often result in representatives with strongly different characteristics belonging to the same category. To address this challenge, a quantitative data-driven classification methodology that leverages machine learningâspecifically, clusteringâis introduced here to achieve unbiased grouping of analyzed consensus protocols implemented in various platforms. When different clustering techniques were used on the analyzed DLT dataset, an average consistency of 78% was achieved, while some instances exhibited a match of 100%, and the lowest consistency observed was 55%.
Federated learning (FL), as an effective method to solve the problem of âdata islandâ, has become one of the hot and widespread concern topics in recent years. However, with the using of FL technology in the practical applications, an increasing number of FL tasks make the training management be more complex and the trade-off of multi-task becomes difficult. To overcome this weakness, this work proposes a privacy-preserving FL framework with multi-tasks using partitioned blockchain, which can run several different FL tasks by multiple requesters. First, a temporary committee is formed for an FL task to facilitating visualization, organization and management of security aggregation. Second, the proposed framework combines Paillier homomorphic encryption with Pearson correlation coefficient to protect users' privacy and ensure the accuracy of global model. Finally, a new blockchain-based reward method is presented to inspire participants to share their valuable data. The experimental results show that the global model accuracy of our proposed framework is able to reach 98.43%. Obviously, the proposed framework is more suitable for practical application environment, especially in industrial application field.
Federated Learning (FL) has recently arisen as a revolutionary approach to collaborative training Machine Learning models. According to this novel framework, multiple participants train a global model collaboratively, coordinating with a central aggregator without sharing their local data. As FL gains popularity in diverse domains, security, and privacy concerns arise due to the distributed nature of this solution. Therefore, integrating this strategy with Blockchain technology has been consolidated as a preferred choice to ensure the privacy and security of participants. This paper explores the research efforts carried out by the scientific community to define privacy solutions in scenarios adopting Blockchain-Enabled FL. It comprehensively summarizes the background related to FL and Blockchain, evaluates existing architectures for their integration, and the primary attacks and possible countermeasures to guarantee privacy in this setting. Finally, it reviews the main application scenarios where Blockchain-Enabled FL approaches have been proficiently applied. This survey can help academia and industry practitioners understand which theories and techniques exist to improve the performance of FL through Blockchain to preserve privacy and which are the main challenges and future directions in this novel and still under-explored context. We believe this work provides a novel contribution respect to the previous surveys and is a valuable tool to explore the current landscape, understand perspectives, and pave the way for advancements or improvements in this amalgamation of Blockchain and Federated Learning.
James Kolapo Oladele, Arnold Adimabua Ojugo, Christopher Chukwufunaya Odiakaose, Frances Uchechukwu Emordi ¡ 8 authors
Blockchain platforms propagate into every facet, including managing medical services with professional and patient-centered applications. With its sensitive nature, record privacy has become imminent with medical services for patient diagnosis and treatments. The nature of medical records has continued to necessitate their availability, reachability, accessibility, security, mobility, and confidentiality. Challenges to these include authorized transfer of patient records on referral, security across platforms, content diversity, platform interoperability, etc. These, are today â demystified with blockchain-based apps, which proffers platform/application services to achieve data features associated with the nature of the records. We use a permissioned-blockchain for healthcare record management. Our choice of permission mode with a hyper-fabric ledger that uses a world-state on a peer-to-peer chain â is that its smart contracts do not require a complex algorithm to yield controlled transparency for users. Its actors include patients, practitioners, and health-related officers as users to create, retrieve, and store patient medical records and aid interoperability. With a population of 500, the system yields a transaction (query and https) response time of 0.56 seconds and 0.42 seconds, respectively. To cater to platform scalability and accessibility, the system yielded 0.78 seconds and 063 seconds, respectively, for 2500 users.
Federated learning-based medical data privacy sharing can promote the development of medical industry intelligence, but limited by its own security and privacy deficiencies, federated learning still suffers from a single point of failure and privacy leakage of intermediate parameters. To address these problems, this paper proposes a privacy protection framework for medical data based on blockchain and cross-silo federated learning, using cross-silo federated learning to establish a collaborative training platform for multiple medical institutions to enhance the privacy of medical data, introducing blockchain and smart contracts to realize decentralized federated learning to enhance trust between distrustful medical institutions and solve the problem of a single point of failure. In addition, a secure aggregation scheme is designed using threshold homomorphic encryption to prevent the privacy leakage problem during parameter transmission. The experimental and analytical results show that the accuracy of this paperâs scheme is consistent with the original federated learning scheme, effectively deals with the problems of single-point failure and inference attacks of federated learning, improves system robustness, and is suitable for medical scenarios with more stringent requirements on security and accuracy.
Federated learning (FL) has shown promise in smart industries as a means of training machine-learning models while preserving privacy. However, it contradicts FLâs low communication latency requirement to rely on the cloud to transmit information with data owners in model training tasks. Furthermore, data owners may not be willing to contribute their resources for free. To address this, we propose a single contract to dual contract approach to incentivize both model owners and workers to participate in FL-based machine learning tasks. The single-contract incentivizes model owners to contribute their model parameters, and the dual contract incentivizes workers to use their latest data to participate in the training task. The latest data draw out the trade-off between data quantity and data update frequency. Performance evaluation shows that our dual contract satisfies different preferences for data quantity and update frequency, and validates that the proposed incentive mechanism is incentive compatible and flexible.
Lu Zhou, Abebe Diro, Akanksha Saini, Shahriar Kaisar ¡ 5 authors
Identity sharing systems, regardless of their architectural models, share common vulnerabilities. These systems compel users to divulge personal information and furnish proof of identity for accessing services, leaving them susceptible to data breaches that can culminate in identity theft and jeopardize online data security. While blockchain technology offers a potential remedy, delivering enhanced security, immutability, and traceability, it simultaneously raises pertinent concerns surrounding privacy and transparency. The integration of zero-knowledge proof (ZKP) technology has emerged as a promising solution, particularly in enhancing privacy within the transparent blockchain ecosystem. Our paper conducts an exhaustive survey of the existing literature, with a particular focus on the assimilation of ZKP technology into blockchain for the secure sharing of user identities. We undertake a critical evaluation of the advancements achieved in this domain, pinpoint the formidable challenges that must be confronted, and uncover nascent opportunities for further exploration. Our contribution transcends the realms of mere summarization and analysis; we go a step further by offering recommendations drawn from real-world case studies and delineating future research directions.
Abstract Timed-release encryption (TRE) is a cryptographic primitive that can control the decryption time and has significant application value in time-sensitive scenarios. To solve the reliability issue of nodes in existing TRE anonymous interaction schemes, we propose a blockchain-based TRE protocol for anonymous query time trapdoors. In our protocol, the recipient divides the encrypted trapdoor request information into n ciphertext fragments using secret sharing technology near the decryption time, and employs the idea of onion routing to perform layer-by-layer encryption, creating onion-type data transmitted through middlemen selected from the smart contract. After receiving the ciphertext fragments, the time server integrates them to obtain the trapdoor request information and returns the corresponding time trapdoor to the recipient. This allows the recipient to query any time trapdoor anonymously. Our protocol provides a normative design for the smart contract and specific constraints on the participantsâ behavior. Compared with the related anonymous query trapdoor schemes, our protocol improves the probability of successful queries. Security analysis shows that our protocol can resist release-ahead attack, interruption attack, eavesdropping attack, and replacement attack. Performance analysis shows that our protocol outperforms related protocols regarding anonymity, efficiency, and flexibility, achieving highly efficient anonymous interactions. Finally, we conducted an experiment in the Ethereum Rinkeby test network. For the settings of ciphertext fragment number $$n=3$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>n</mml:mi><mml:mo>=</mml:mo><mml:mn>3</mml:mn></mml:mrow></mml:math> and ciphertext fragment threshold $$t=2$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>t</mml:mi><mml:mo>=</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:math> , the gas consumption for a user to execute the contract was $5.66, which was higher than the contract cost of related schemes, but the contract execution cost was within an acceptable range.
In recent years, Wireless Sensor Networks (WSNs) have become integral in various applications ranging from environmental monitoring to defense. However, the security and reliability of these networks remain a paramount concern due to their susceptibility to various types of cyber-attacks and failures. This paper proposes a novel integration of blockchain technology with WSNs to address these challenges. Blockchain, with its decentralized and tamper-resistant ledger, offers a robust framework to enhance the security and reliability of sensor networks. The study begins by analyzing the current security threats and challenges faced by WSNs, emphasizing the need for a solution that can ensure data integrity, confidentiality, and network resilience. We then introduce blockchain technology and discuss its key features such as decentralization, immutability, and consensus algorithms, which are beneficial in creating a secure and reliable WSN environment. Subsequently, we present a detailed architecture of how blockchain can be integrated with WSNs. This includes the deployment of a lightweight blockchain protocol suited for the limited computational resources of sensor nodes. We also explore the use of smart contracts for automated, secure data handling and network management within WSNs. To validate the proposed integration, we conduct a simulations based on network attacks. The results demonstrate significant improvements in the security and reliability of WSNs when blockchain is implemented. This is evidenced by enhanced resistance to common attacks, such as data manipulation and node compromise and increased network uptime.
As the volume of healthcare and medical data increases from diverse sources, real-world scenarios involving data sharing and collaboration have certain challenges, including the risk of privacy leakage, difficulty in data fusion, low reliability of data storage, low effectiveness of data sharing, etc. To guarantee the service quality of data collaboration, this paper presents a privacy-preserving Healthcare and Medical Data Collaboration Service System combining Blockchain with Federated Learning, termed FL-HMChain. This system is composed of three layers: Data extraction and storage, data management, and data application. Focusing on healthcare and medical data, a healthcare and medical blockchain is constructed to realize data storage, transfer, processing, and access with security, real-time, reliability, and integrity. An improved master node selection consensus mechanism is presented to detect and prevent dishonest behavior, ensuring the overall reliability and trustworthiness of the collaborative model training process. Furthermore, healthcare and medical data collaboration services in real-world scenarios have been discussed and developed. To further validate the performance of FL-HMChain, a Convolutional Neural Network-based Federated Learning (FL-CNN-HMChain) model is investigated for medical image identification. This model achieves better performance compared to the baseline Convolutional Neural Network (CNN), having an average improvement of 4.7% on Area Under Curve (AUC) and 7% on Accuracy (ACC), respectively. Furthermore, the probability of privacy leakage can be effectively reduced by the blockchain-based parameter transfer mechanism in federated learning between local and global models.
Syed M. Aqleem Abbas, Muazzam A. Khan, Wadii Boulila, Anis Kouba ¡ 6 authors
Unmanned aerial vehicles (UAVs) can be used as dronesâ edge Intelligence to assist with data collection, training models, and communication over wireless networks. UAV use for smart cities is rapidly growing in various industries, including tracking and surveillance, military defense, managing healthcare delivery, wireless communications, and more. In traditional machine learning techniques, an enormous amount of sensor data from UAVs must be shared to central storage to perform model training, which poses serious privacy risks and risks of misuse of information. The federated learning technique (FL), which can be applied to UAVs, is a promising means of collaboratively training a global model while retaining local access to sensitive raw data. Despite this, FL is a significant communication burden for battery-constrained UAVs due to local model training and global synchronization frequency. In this article, we address the major challenges associated with UAV-based FL for smart cities, including single-point failure, privacy leakage, scalability, and global model verification. To tackle these challenges, we present a differentially private federated learning framework based on Accumulative Reputation-based Selection (ARS) for the edge-aided UAV network that utilizes blockchains to prevent single-point failures where we switched from central control to decentralized control, Interplanetary File System (IPFS) for off-chain model storage and their respective hash-keys on-chain to ensure model integrity. Due to IPFS, the size of the blockchain will be reduced, and local differential privacy will be applied to prevent privacy leakages. In the proposed framework, an aggregator will be selected based on its ARS score and model verification by the validators. After most validators approve it, it will be available for use. Several parameters are taken into consideration during evaluation, including accuracy, precision, recall, F1-score, and time consumption. It also evaluates the number of edge computers vs test accuracy, the number of edge computers vs time consumption for global model convergence, and the number of rounds vs test accuracy. This is done by considering two benchmark datasets: MNIST and CIFAR-10. The results show that the proposed work preserves privacy while achieving high accuracy. Moreover, it is scalable to accommodate many participants.