Soleti Navya, Pankaj Bharati, S. Sree Hari Raju, Anil Kumar G · 6 authors
Blockchain technology is popular for its safe and decentralized architecture that help build digital contracts, supply chain management, and financial services applications. The key features of this technology is that smart contracts automatically perform agreements when certain predefined conditions are met. Coding errors or design flaws make them vulnerable to security attacks. People with malicious intent may exploit these flaws, resulting in significant financial losses and a decline in trust in blockchain technology as a whole. With help of artificial intelligence and machine learning algorithms this paper identifies anomalies in smart contracts and increase blockchain security by its mitigation. The machine learning algorithm is trained to identify anomalies that deviate from standard norms by analyzing patterns and behavior. The paper focuses on identifying common vulnerabilities that attackers may exploit like overflows, unauthorized access channels and re-entrancy bugs using a gradient boosting classifier. Developers and other stakeholders can be informed about possible security vulnerabilities before they become serious threats by using anomaly detection to generate early alerts. The primary goal of this research is to create a strong and reliable security tool that will significantly improve blockchain security and guarantee that smart contracts operate as intended and increase public confidence in blockchain technology.
The smart grid is the next evolution of electrical power systems, a continuation of the old grids that involves a mix of digital and traditional power grid technologies to allow the potential to communicate in both directions, decentralized energy production and real-time monitoring. However, such a connection exposes it to cyber attacks, data fraud, and unauthorized access as well. Blockchain technology is one of these technologies because it is transparent, immutable, and decentralized to overcome these security obstacles. In this paper, an overview of blockchain technology smart grid security, architecture, consensus algorithm, and application are presented. Some of the most notable blockchain works in the smart grid include secure energy trading, decentralised identity management, detecting attacks and preserving privacy. When applied in smart grids, reviewed blockchain protocols also comprise Proof of Work (PoW) and Proof of Stake (PoS) along with Practical Byzantine Fault Tolerance (PBFT). Top of that, there are hybrid types of blockchain such as artificial intelligence (AI) and the Internet of things (IoT) that are also covered as the next picture to enable the system to become more scalable and interoperable. Power consumption, time wastage, and regulation hurdle is greatly considered. This paper has concluded that blockchain is a bottom-up technology, which can cause smart grid infrastructures to be much more resilient, transparent, and efficient.
Bitcoin's limited scripting capabilities and lack of native interoperability mechanisms have constrained its integration into the broader blockchain ecosystem, especially decentralized finance (DeFi) and multi-chain applications. This paper presents a comprehensive taxonomy of Bitcoin cross-chain bridge protocols, systematically analyzing their trust assumptions, performance characteristics, and applicability to the Artificial Intelligence of Things (AIoT) scenarios. We categorize bridge designs into three main types: naive token swapping, pegged-asset bridges, and arbitrary-message bridges. Each category is evaluated across key metrics such as trust model, latency, capital efficiency, and DeFi composability. Emerging innovations like BitVM and recursive sidechains are highlighted for their potential to enable secure, scalable, and programmable Bitcoin interoperability. Furthermore, we explore practical use cases of cross-chain bridges in AIoT applications, including decentralized energy trading, healthcare data integration, and supply chain automation. This taxonomy provides a foundational framework for researchers and practitioners seeking to design secure and efficient cross-chain infrastructures in AIoT systems.
The rapid expansion of the Internet of Things (IoT) is driving the integration of billions of connected devices across various domains, including healthcare, transportation, and smart urban systems. Although this proliferation offers considerable advantages in terms of functionality and operational efficiency, it also brings to the forefront a range of pressing concerns, particularly in relation to security, reliability, and privacy. These challenges are largely rooted in the decentralized and dynamic architecture of IoT ecosystems. In this context, trust and reputation mechanisms have become increasingly vital for enabling secure and reliable interactions between devices and users. This paper examines recent advances in trust management models tailored to IoT environments, with a focus on approaches leveraging blockchain technologies, machine learning techniques, and edge or fog computing paradigms. We assess the practical implications of these solutions, discussing both their strengths and inherent limitations. Furthermore, we identify key open issues such as scalability, data protection, and interoperability across platforms, and we outline potential research directions to support the development of more robust and adaptable trust frameworks for the evolving IoT landscape.
Sep 4, 2025·2025 IEEE 13th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS)
Blockchain technology is increasingly adopted in Internet of Things (IoT) logistics to enhance data security, scalability, and real-time transparency. The security threats and data silos in IoT logistics networks forces the adoption of blockchain technology since its tamper-proof distributed ledger system provides secure and scalable tracking capabilities. Blockchain-based system performance and operational suitability for time-sensitive operations depends heavily on the chosen consensus algorithm. The research evaluates five blockchain consensus methods (Proof of Work (PoW), Proof of Stake (PoS), Practical Byzantine Fault Tolerance (PBFT), Raft, and HotStuff) for IoT logistics operations. The research utilized Ethereum for PoW/PoS testing and Hyperledger Fabric for PBFT/Raft and HotStuff prototype testing and measured throughput and latency through standard tools. The results demonstrate that traditional Proof of Work and Proof of Stake systems provide poor performance in terms of transaction speed and high latency that makes them unusable for real-time data processing requirements. In contrast, the permissioned consensus algorithms PBFT, Raft and HotStuff demonstrate higher transaction rates and faster confirmation times compared to traditional mechanisms. HotStuff provides stronger fault tolerance and scalable performance compared to traditional PBFT. Raft demonstrates the highest performance in normal operating conditions. The research demonstrates that consortium or private blockchains using Byzantine fault-tolerant or crash fault-tolerant consensus within restricted environments are better suited for realtime logistics IoT systems than public blockchains based on PoW/PoS.
There has been a growing need for secure, transparent, and efficient systems to manage agricultural processes. Traditional methods often lack traceability, leading to inefficiencies, fraud, and data manipulation. Additionally, centralized agricultural management systems are prone to security risks and single points of failure. Blockchain technology provides a decentralized and secure solution that enhances transparency, security, and efficiency in agriculture. By utilizing a distributed ledger, blockchain ensures end-to-end traceability of agricultural products, preventing fraud and ensuring quality control. Farmers, suppliers, and consumers can verify the authenticity of products without relying on intermediaries, thereby reducing costs and enhancing trust in the supply chain using the proposed scheme.
Jonas Lopes de Vilas Boas, Ygor S. Costa, Rodrigo da Rosa Righi, Antônio Marcos Alberti · 5 authors
Reliable vaccine tracking and monitoring during transport and storage are essential to ensure dose effectiveness while minimizing waste. However, current solutions face challenges related to reliability, immutability, security, transparency, flexibility, extensibility, patient support, trust, and cost. Centralized systems are vulnerable to fraud, tampering, and manipulation, often relying on manual service contracts and lack of attested IoT devices to ensure data authenticity. Moreover, most existing platforms do not provide tamper-proof, near real-time monitoring, resulting in operational vulnerabilities and increased costs. This article presents Coldnet, a novel architecture for vaccine tracking and tracing that addresses these issues by: (i) integrating IoT device attestation with the registration of immutable data and flexible monitoring attributes; (ii) using Blockchain-based smart contracts to automatically manage tracking and monitoring clauses, improving security and enabling dynamic rule management; (iii) offering intuitive interfaces to support patient access to delivery information; and (iv) deploying an affordable, user-friendly IoT prototype to monitor and report vaccine status. A case study demonstrates Coldnet's feasibility, with an average delay of 20 seconds for recording and checking conditions — suitable for real operations. A simulation evaluating scalability and the impact of IoT attestation shows transaction costs of US$1.51 for ten vaccine batches with five monitored properties each, a cost deemed acceptable for the added features. Execution delays remained stable (0.85–0.92 seconds), with negligible impact from attestation. Coldnet contributes to reliable vaccine logistics, improving public health efforts by strengthening trust, transparency, and data integrity in vaccination campaigns.
In recent times., the accelerated growth of data has established big data analytics as an integral part of decision making in various industries., including healthcare., finance., manufacturing., retail etc. With the rise of big data., cloud computing., the relatively recent trend of using remote servers to hold and process data., has become a powerful tool for managing large datasets., providing storage and computational capabilities that are scalable., economical., and also elastic. But., when data is getting complex and distributed., security., privacy and data integrity have emerged as major concerns. The inherent features of blockchain technology like decentralization., immutability., and transparency can serve as a revolutionary alternative to tackle these challenges., while improving the efficiency and trustworthiness of big data analytics based on cloud. Data integrity., tampering prevention., and transparency-based auditing mechanisms help to secure data on cloud systems., and blockchain can improve these aspects. Additionally., the distributed nature of blockchain makes it possible to securely share data across different organizations without sacrificing privacy., which is especially important in industries like healthcare and finance. In addition., incorporating blockchain technology enables efficient tracking of data provenance., guaranteeing the verifiability of data origins and transformations., thereby enhancing trust and transparency. Blockchain has applied the security enhancement of cloud-based big data analytics platforms., which can also improve processing efficiency. Similarly., various consensus mechanisms used by blockchain i.e.., Proof of Work., Proof of Stake etc.., aid in cutting down the amount of time and resources required to validate and authenticate data in real-time., thus making a faster yet more secure processing time possible. Moreover., self-executing contracts (smart contracts) define terms that are directly built into the blockchain., and can automate data workflows and decisions., thus improving efficiencies. By ushering in new efficiencies that could sharply trim operational budgets and accelerate data- based decision-making across an array of sectors., these abilities hold immense potential.
This research presents a decentralized platform for academic credential verification using blockchain technology. The system allows students to store and share their school diplomas in a secure, tamper-evident format. By leveraging smart contracts, diploma issuance and verification are automated, enabling educational qualifications to be independently verified without manual intervention. The proposed system significantly reduces verification time, requiring only 400 milliseconds for an individual diploma and less than 2 seconds for a complete document. These performance metrics demonstrate the system's efficiency in real-world use. Employers can validate job applicants' credentials directly through the blockchain, reducing administrative workload and improving trust in academic records. The solution benefits students, educational institutions, and employers alike by ensuring the authenticity and integrity of issued diplomas through a decentralized and transparent infrastructure.
The research proposes and evaluates a blockchain-based AIoT inspection system implemented on an NVIDIA Jetson Xavier NX edge computing device. The system integrates AI-powered inspection capabilities with Hyperledger Fabric blockchain technology to ensure tamper-proof record-keeping and data integrity. Performance analysis demonstrates the feasibility of running distributed ledger technology in conjunction with AI processing on edge computing devices. The implementation achieves reliable verification of inspection data while maintaining acceptable computational resource usage and energy consumption. Results indicate that blockchain integration prevents unauthorized data manipulation without significantly compromising system performance.
Thi Thanh Thuy, Minh‐Ky Nguyen, Thuyet D. Bui, Hoang Phan Hải Yen · 10 authors
This paper explores how blockchain technology, widely known as the backbone of cryptocurrencies, can be harnessed to address limitations of traditional water quality monitoring (WQM) systems. Blockchain offers a decentralized, tamper-proof ledger that enables secure, transparent, and traceable data management across distributed networks. When applied to water quality monitoring, blockchain facilitates real-time data acquisition, enhances data integrity, and enables smart contracts for automated regulatory compliance and alerts. These features not only improve the accuracy and efficiency of WQM systems but also build public trust in the reported data. Key insights from current research and pilot applications highlight blockchain’s capacity to integrate with IoT devices for real-time sensing, support adaptive water governance, and empower local stakeholders through decentralized control and transparent access to information. The implications for policy and practice are significant: blockchain-based WQM can support stronger regulatory enforcement, encourage cross-sector collaboration, and provide a robust digital foundation for sustainable water management in smart cities and rural areas alike. As such, this review paper positions blockchain as a transformative tool in the digital transition toward more resilient and equitable water management systems.
Anurag Shrivastava, RVS Praveen, Mohammed I. Habelalmateen, Arti Badhoutiya · 6 authors
The integration of blockchain technology with the Internet of Things (IoT) and Industrial Internet of Things (IIoT) heralds a transformative era for numerous industries by enhancing security, ensuring data integrity, and fostering trust through decentralized mechanisms. This comprehensive survey explores the convergence of blockchain solutions within IoT and IIoT frameworks, highlighting the technological advancements, security challenges, and potential attacks that characterize this evolving landscape. By scrutinizing various blockchain architectures, consensus mechanisms, and real-world applications alongside a critical examination of security threats including DoS, fuzzers, and combined assaults, the study offers an in-depth understanding of both the potential and the pitfalls of blockchain in IoT and IIoT contexts. Through analyzing performance metrics against attacks and discussing mitigation strategies, this survey aims to provide a roadmap for researchers, practitioners, and policymakers to navigate the complexities of securing blockchain-enabled IoT and IIoT systems.
Harsh Yadav, Varun Shukla, Hamdan Zaman khan, Anshul Kumar Mandal
The currently existing Aadhaar data sharing system has a dearth of granular user control enabling other third party entities including banks, telecom service providers, and government agencies access to the user's data without explicit, time bound and purpose specific consent. This incites all manner of privacy, transparency, and data misuse concerns. This paper then proposes a Blockchain Based Consent as a Service (CaaS) Framework to guarantee such real time, secure and auditable access of Aadhaar linked data in the hands of users via authorized smart contracts. In the proposed model, the smart contract encapsulates each data access request includes purpose, requesting entity, the type of data, and the duration of access. A secure app or web interface will notify user or browse regarding consent requests, and allow him to accept or refuse request by using Aadhaar Virtual ID and OTP or digital signatures. Once it is approved, the access is granted to consumers on a temporary basis with full logging on a blockchain ledger that is immutable and open for all to see by users and regulators. It is compliant with the Digital Personal Data Protection (DPDP) Act 2023 and utilizes off chain storage of encrypted data. In addition to allowing users to possess granular control over their digital identity, the framework also lays the groundwork for Zero Knowledge Proof (ZKP) based functionalities and AI based anomaly detection to proactively protect users' privacy in the future.
Mohamad Abdul Kadar, Suthari Yugandhar Reddy, Prakruti Parmar, Jobanpreet Kaur · 6 authors
The expansion of critical applications using distributed systems has established significant worries about protection of data and security throughout networks. Conventional protective methods struggle to gain control of decentralized network environments because trust and control distribution occurs among multiple participants. This study investigates how blockchain technology with machine learning approaches creates better data protection and system defense in distributed environments. Through blockchain technology users gain an unchangeable open ledger system which maintains data purity and through machine learning they get sophisticated analytic methods for spotting irregularities and forecasting security threats coupled with optimal security enhancement. The research examines existing approaches followed by a framework description for blockchain integration and their dual effects on security protocol enhancement. The paper demonstrates how blockchain and machine learning combine to solve modern distributed system security requirements as well as privacy protection and safe sharing in financial and healthcare sectors and IoT systems through an extensive assessment. The last part examines both research obstacles and contractions for the developing sector of study.
Securing and protecting privacy of patient records are crucial challenge in block chain. In this paper, proposed Secure Hybrid Framework for Blockchain(SHFB) is confronted by integrating AES-GCM and ECC with Zero Knowledge Proofs and Fully Homomorphic Encryption (FHE) to enhance the security efficiency. The exponential growth of healthcare data necessitates secure and efficient frameworks to handle real time processing and protection against cyber threats. This paper presents the SHFB (Confronted) model, a block chain integrated architecture optimized for high throughput, robust denial of service (DoS) resistance, and low latency performance in electronic health record (EHR) systems. Comparative analysis reveals that SHFB surpasses conventional models such as Decentralized Security, AI Privacy and Secure EHR in key metrics, making it a scalable and secure solution for modern healthcare infrastructure. Patient records were used to conduct experimental evaluation and demonstrate that SHFB significantly reached high to compare with existing models. Particularly SHFB achieves an encryption time of 0.02 seconds, security efficiency of 95%.The proposed method shows the effectiveness of hybrid method by enhancing the performance and security of block chain in management of data using healthcare.
Remote service delivery and automation using Blockchain and Internet of Things (IoT) are revolutionising healthcare operations. Enhancing healthcare data interchange and providing real-time treatment is becoming more difficult due to the exponential growth of patient populations around the globe. It is still very difficult to develop a digital healthcare platform that is entirely decentralised, secure, trustworthy, interoperable, and scalable, even if existing studies have improved these platforms to improve patient outcomes and reduce hospital visits. This paper proposes a robust and scalable healthcare architecture using Blockchain smart contracts and IoT, with the integration of the InterPlanetary File System (IPFS). By securely storing sensitive medical records, the framework enhances data privacy and interoperability for patients as well as healthcare professionals. A device proxy monitors potentially vulnerable IoT devices and uses cryptography to ensure that data remains private. Experimental evaluations of the system’s performance have focused on key factors, including healthcare record upload, download, access, and mining times. The results show that public healthcare systems based on the Blockchain considerably boost efficiency and performance by integrating IPFS.
This article explores how the integration of Artificial Intelligence (AI), Machine Learning (ML), Web3.0, Blockchain, Metaverse, and Non-Fungible Tokens (NFTs) will revolutionize various aspects of public life globally over the next decade. We introduce novel perspectives such as AI-driven decentralized governance, blockchain-based universal basic income, and metaverse-enabled global education platforms. These technologies will transform global supply chains through AI-driven forecasting and blockchain-verified logistics, ensuring transparency and efficiency. Healthcare will advance with AI-powered telemedicine and personalized treatments, reducing disparities in underserved regions. Autonomous systems will enhance urban mobility and disaster response, fostering sustainable smart cities. Web3.0 will empower users with decentralized digital identities and data sovereignty, redefining advertising and social media through token-based models. Blockchain will secure academic credentials, streamline insurance, and enable transparent philanthropy, while carbon credit markets promote sustainability. The metaverse will revolutionize remote work, healthcare consultations, and cultural preservation through immersive virtual environments. NFTs will democratize real estate and creative economies, enabling tokenized ownership and secure voting systems. Synergistically, these technologies will create decentralized e-commerce, disaster response systems, and virtual innovation hubs, fostering equitable digital ecosystems. However, challenges like digital divides, AI biases, and blockchain scalability must be addressed to ensure inclusive adoption. This article envisions a future where these advancements redefine governance, economies, and social interactions, paving the way for an innovative, equitable global society. AI-driven avatars and decentralized AI training platforms will further enhance virtual collaboration, while tokenized cultural assets empower communities, ensuring a resilient, inclusive digital future.
Farooq Ahmed, Teng Zhou, Hazrat Bilal, Faiz Ul Islam · 6 authors
The convergence of blockchain with Industry 5.0 technologies presents significant opportunities for healthcare data management; however, current systems face challenges related to scalability, privacy, and energy efficiency. This article introduces an innovative framework that leverages ciphertext-policy attribute-based encryption (CP-ABE), Ethereum smart contracts, and decentralized IPFS storage to address these challenges. The framework presents three main innovations: 1) a human-centric authentication system that ensures security without sacrificing cryptographic integrity; 2) post-quantum Kyber-786 algorithms paired with CP-ABE, which minimizes computational overhead by 27% while facilitating 32 ms key generation; and 3) an energy-efficient Proof of Stake (PoS) consensus mechanism that reduces energy consumption by 98% (0.05 kW/transaction) compared to traditional blockchain systems. Thorough testing demonstrates 99.5% resistance to man-in-the-middle attacks, a throughput of 15.6 MB/min at scale, and an emergency access latency of under 120 ms, which is essential for practical healthcare applications. By consolidating decentralized pseudo-identities for patient anonymity, secure audit logs, and GDPR/HIPAA-compliant data governance, this research establishes a new standard for secure, sustainable, and patient-focused health data ecosystems in the Industry 5.0 landscape.
D. Jeevitha, P. Venkadesh, G. Bharathi, N. J. P. Subhashini · 6 authors
Counterfeit products remain a pervasive challenge, undermining consumer trust and impacting global supply chains. In the pharmaceutical and healthcare sectors, counterfeit drugs pose a serious threat to patient safety, leading to ineffective treatments, health complications, and even fatalities. The World Health Organization (WHO) estimates that $10 \%-30 \%$ of medicines in developing countries are counterfeit [1], highlighting the urgent need for a secure verification mechanism. Leveraging blockchain technology integrated with QR codes & NFT tags offers a powerful solution for authentic product identification. This study proposes a decentralized framework where unique QR codes, linked to blockchain records & Non-fungible tokens, enable end-to-end traceability and secure verification of product authenticity. Each product is assigned a unique digital identity recorded on an immutable blockchain ledger, accessible through a simple QR code scan linked to an NFT token that shows the ownership details of the product. The system enhances transparency, empowers consumers to verify product authenticity, and strengthens brand protection. The integration of NFTs with QR codes enhances authentication accuracy by up to $85 \%$, reducing the risk of counterfeit drugs entering the supply chain [2]. Unlike traditional QR codes, which can be duplicated, NFTs provide a unique, tamper-proof digital certificate that verifies product legitimacy, ensuring a higher level of security. This paper explores the technical implementation of the proposed solution, evaluates its application across industries, and addresses challenges such as scalability, user adoption, and cost efficiency. The findings demonstrate that blockchain, NFT, and QR code integration can significantly enhance product authentication, creating a safer and more trustworthy marketplace.
A decentralized crowdfunding system using Ethereum blockchain technology is proposed to resolve problems within traditional crowdfunding portals, such as a lack of transparency, high dependence on intermediaries, and poor supervision from donors. The proposed system is capable of automating refund mechanisms as well as milestone-based fund disbursements while maintaining transparent transaction logs through the use Solidity-based smart contract. Unlike the traditional models where funds are kept and controlled by custodians and managed by third parties, this system enables participants to control the release of funds based on completion of predefined project stages. The system was implemented using Solidity, React.js, Node.js, and deployed on Sepolia Ethereum Testnet. Etherscan was used for transaction validation, and MetaMask for user wallet interactions, enabling successful campaign creation, secure contribution submission, and decentralized fund flow. Limitations include logic flaws in post-deadline countdown timers and a focus on a single tokens type. Potential enhancements include automated refunds for failed campaigns, fiat currency on-ramps, and fund control via DAO voting. The proposed framework focuses on solving practicality, security, and efficiency issues in existing crowdfunding models using blockchain technology.
Water usage serves as a key factor in boosting groundwater levels, alleviating water scarcity, and enhancing sustainable farming practices. Yet, ineffective oversight and accountability among stakeholders often undermine water resource management (WRM). To address this issue, we introduce a framework called S2WRM, which integrates a dual strategy: a game theory-based method and a blockchain-based system. In the game theory component, we apply a Stackelberg game, positioning the Water Resource Management Authority (WRMA) as the leader and industrialists and farmers as followers. Through this sequential game, dynamic pricing is determined as the leader establishes pricing strategies across various scenarios, analyzing usage trends to boost WRMA revenue. In addition, a blockchain incentive system is implemented, using ERC-20 contracts to mint and distribute tokens as rewards. This blockchain framework is built on the Ethereum platform, with Stackelberg simulations coded in Python. Analysis of different scenarios reveals key insights: first, blockchain-based incentives improve consumption oversight, fostering transparency in stakeholder supply chains second, WRMA revenue rises as blockchain reduces overhead costs tied to dynamic pricing and third, sustainability is advanced through token limits and penalties.
RÉSUMÉ: Au cours de la dernière décennie, la technologie blockchain a rapidement évolué d’une in-novation émergente à un pilier fondamental de l’infrastructure numérique moderne, susci-tant un vif intérêt tant dans le monde académique qu’industriel. Son adoption en tant que composante essentielle des systèmes financiers, de l’Internet des objets (IoT), de l’Internet industriel des objets (IIoT), des environnements Metaverse et des plateformes de contrats intelligents s’explique par ses propriétés intrinsèques de transparence, d’immutabilité et de décentralisation. Toutefois, la croissance exponentielle des applications blockchain s’est ac-compagnée de l’apparition de défis de sécurité sophistiqués, menaçant la stabilité et la fiabilité de ces systèmes. Parmi les problèmes les plus critiques figurent diverses attaques visant à la fois les couches réseau et applicative. Les systèmes fondés sur la blockchain deviennent de plus en plus vulnérables à des menaces de niveau réseau telles que l’attaque Eclipse, le minage égoïste (selfish mining) et les attaques Sybil. Les attaquants ont démontré leur capacité à lancer ces attaques pour perturber les services, compromettre l’intégrité des données ou prendre un contrôle disproportionné du réseau. Au niveau applicatif, les vulnérabilités au sein des con-trats intelligents—telles que la réentrance, les dépassements et sous-dépassements d’entiers (overflow/underflow), ou la dépendance aux horodatages—constituent des failles majeures que les adversaires peuvent exploiter afin de manipuler la logique contractuelle, détourner des fonds ou subvertir le fonctionnement attendu. Pour répondre à ces problématiques multidimensionnelles, cette thèse propose un ensemble de cadres de sécurité complets, chacun étant adapté à une classe spécifique de menaces et couvrant divers types de blockchains (y compris Bitcoin et Ethereum) ainsi que différents mécanismes de consensus (preuve de travail, preuve d’enjeu, preuve d’autorité). La recherche catégorise systématiquement les stratégies de défense en trois axes principaux : la détection d’intrusion, la mitigation des attaques et la prévention des menaces. Plus précisément, la thèse présente deux nouvelles approches de détection, une technique de mitigation et deux stratégies de prévention. Au niveau réseau, la première composante vise les attaques Eclipse dans les systèmes Meta-verse basés sur la blockchain. Le Metaverse—domaine numérique émergent dédié aux activ-ités sociales, éducatives et commerciales—repose fortement sur la blockchain pour la gestion sécurisée des actifs et le contrôle décentralisé. Néanmoins, il demeure vulnérable à des at-taques de grande ampleur, notamment les attaques Eclipse, susceptibles d’isoler des nœuds et de faciliter d’autres exploits tels que le selfish mining ou la double dépense. Cette étude introduit un cadre de détection d’intrusion s’appuyant sur des algorithmes de détection de communautés, en particulier l’algorithme de Leiden, afin d’identifier avec précision les nœuds malveillants du réseau. La deuxième composante s’intéresse au minage égoïste, ou selfish mining, une menace pertur-batrice qui compromet l’intégrité du consensus des blockchains majeures telles que Bitcoin. Cette thèse propose une solution globale couvrant à la fois la détection et la mitigation. Le cadre de détection utilise des ensembles de données simulés du réseau, une sélection avancée de caractéristiques et une classification par forêts aléatoires pour identifier efficacement les comportements malveillants de minage. Pour la mitigation, la thèse développe une taxonomie détaillée des approches basées sur la théorie des jeux, applicables à la sécurité de l’IoT sur blockchain, facilitant ainsi le choix de modèles pertinents pour analyser le comportement des acteurs égoïstes. En particulier, un jeu de la guerre d’usure (war of attrition) est utilisé, où des mécanismes de récompense et de punition orientent les stratégies des mineurs, et la théorie des jeux évolutionnistes permet d’analyser la convergence et la stabilité évolutive de la participation honnête au sein du réseau. ABSTRACT: Over the past decade, blockchain technology has rapidly advanced from a novel innovation to a cornerstone of modern digital infrastructure, garnering significant attention from both academia and industry. Its adoption as a core component in financial systems, Internet of Things (IoT), Industrial IoT (IIoT), metaverse environments, and smart contract platforms has been driven by its intrinsic properties of transparency, immutability, and decentraliza-tion. However, the exponential growth of blockchain applications has been paralleled by the emergence of sophisticated security challenges that threaten the stability and reliability of these systems. Among the most critical issues are a range of security attacks targeting both network and application layers. Blockchain-based systems are increasingly vulnerable to network-level threats, such as eclipse, selfish mining, and Sybil attacks. Attackers have demonstrated the ability to launch these attacks to disrupt services, compromise data integrity, or gain dis-proportionate control over the network. At the application layer, code-level vulnerabilities within smart contracts—such as reentrancy, integer overflow and underflow, and timestamp dependency—constitute critical security weaknesses that adversaries can exploit to manipu-late contract logic, drain funds, or subvert intended operations. To address these multifaceted problems, this thesis proposes several security solutions, each tailored to a specific class of threats and spanning various blockchain types (including Bitcoin and Ethereum) and consensus mechanisms (Proof of Work, Proof of Stake, and Proof of Authority). The research systematically categorizes defense strategies into three primary areas: intrusion detection, attack mitigation, and threat prevention. Specifically, the thesis presents two novel detection approaches, one attack mitigation technique, and two strategies for threat prevention. At the network level, the first component targets eclipse attacks in blockchain-based meta-verse systems. The Metaverse—an emerging digital domain for social, educational, and commercial activities—relies heavily on blockchain for secure asset management and decen-tralized control. Nevertheless, it remains susceptible to large-scale security attacks, notably eclipse attacks, which can isolate nodes and facilitate further exploits such as selfish min-ing and double-spending attacks. This study introduces an intrusion detection framework leveraging community detection algorithms, particularly the Leiden algorithm, to accurately identify malicious nodes in the network. The second component addresses selfish mining, a disruptive threat that undermines the consensus integrity of prominent blockchains such as Bitcoin. This thesis presents a com-prehensive solution encompassing both detection and mitigation strategies. The detection framework leverages simulated network datasets, advanced feature selection, and random forest classification to efficiently and accurately identify malicious mining behaviors. For mitigation, the thesis develops a detailed taxonomy of game-theoretic approaches applicable to blockchain-based IoT security, facilitating the selection of appropriate models for analyzing the behavior of selfish actors. In particular, a war of attrition game is employed, with reward and punishment mechanisms guiding miner strategies, and the evolutionary game theory is then used to analyze the convergence and evolutionary stability of honest participation within the network.