This thesis examines three distinct topics on settlement microstructure and market efficiency in both DeFi and TradFi. Chapter one introduces the thesis’ unifying lens, arguing that settlement microstructure drives market efficiency across these markets. It links the three papers by showing how access in Bitcoin private channels, timing in Ethereum intertemporal gas hedging, and composition in equity market retail participation jointly determine fees, latency, liquidity, and price discovery, while previewing the policy framework that renders these mechanisms legible, bounded, and measurable. Chapter two, based on the working paper “Private Settlement in Blockchain Systems” with Dr. Alfred Lehar, provides evidence that the settlement market in blockchain systems is not purely transactional and diverges from the predictions of a simple competitive auction model. Using data from the Bitcoin blockchain, we find that 5.88% of transactions, labeled as private, bypass the competitive auction and are routed directly to miners. Despite being more active than the average user, these transactions are consistently confirmed by a single miner, a statistically unlikely outcome in a competitive environment. Our findings suggest that high-demand users form long-term agreements with miners, paying, on average, 20% lower fees. This chapter also documents how such settlement contracts are structured and operate within an unregulated market. Chapter three, based on the working paper “Gas Tokens: Market for Future Settlement in the Ethereum Blockchain” with Dr. Alfred Lehar, examines the implications of gas tokens as a potential market for future settlement within the Ethereum network. We show that sophisticated and frequent users are more engaged in gas token markets, pre-purchasing tokens to hedge against fluctuations in gas prices and paying, on average, 15.25% lower settlement fees. Moreover, bots actively pursue arbitrage opportunities in gas token markets and hold substantial volumes. Our findings indicate that traded gas token prices have strong predictive power for future gas prices. This research contributes to the development of modern financial instruments for price discovery and hedging within the Ethereum network as a two-sided market. We also empirically analyze the implementation of the Ethereum Improvement Proposal EIP-1559 as a natural experiment. Chapter four, based on my working paper “Silencing the Noise: Amplified Effects, A Causal Study on Price Efficiency”, investigates the causal effects of noise trader removal on market liquidity. In September 2022, an unexpected internet disruption in Iran restricted noise traders while informed traders retained access through brokers. This disruption led to a 6.65-fold increase in the bid-ask spread and a 46.8% decrease in informed trade speed due to market access asymmetry. Social media censorship in affected regions further amplified information asymmetry, resulting in a 7.2% price impact. Using a five-year analysis of political unrest, this study disentangles the effects of unrest and internet disruption on noise trading activity. The findings reveal that political unrest increases regional noise trading activity, whereas internet disruption decreases it. When both unrest and internet disruption occur simultaneously, regional noise trading activity decreases by 23.5%. This paper provides novel insights into market microstructure and the dynamics of liquidity provision through noise trading in emerging markets.
Surenthran David, S. Aravinth, Chris Sherin D, Mohamed Issath S · 6 authors
The banking sector is disrupted by the development of Decentralized Finance (DeFi) based on blockchain technology. DeFi provides permissionless, trustless, and programmable services. This article looks at how DeFi is changing the traditional banking sector from a blockchain viewpoint. DeFi both threatens and offers opportunities to conventional banks by eliminating middlemen, lowering costs of transactions, improving financial inclusion, and facilitating real-time asset settlement. For this research, we use a unique approach combining financial network modeling and smart contract analysis to measure the efficiency, security, and scalability of DeFi protocols compared to centralized banking systems. While the findings suggest that DeFi could be more accessible, efficient, and open than existing financial infrastructures, they also show that there are real problems with regulation, security, and systemic risk. This work adds to our knowledge of the changing dynamics between traditional finance and DeFi by providing a blockchain-based model for the potential interplay between these two paradigms in the future.
This research suggests a groundbreaking methodology for the optimization of Supply Chain Finance (SCF) by integrating Deep Belief Networks (DBN) and Proof of Stake (PoS) blockchain techniques. The research collects business details, financial tactics, and supplier information from the Kaggle dataset. This paper uses Tokenization for the process of preprocessing, and it uses Locally Linear Embedding (LLE) method to reduce the dimensionality and in the protection of limited structures. Then, the optimization is done by DBNs, which will improve the method’s accuracy, whereas the PoS secures the data with encryption, decryption, and hashing. Moreover, the method’s effectiveness is evaluated using performance analyses such as computational time, efficiency ratio, error ratio, data authentication, and data management ratio. The proposed DBN-PoS is then compared with some existing methods like Self-adaptive Tasmanian Devil Optimization (SA-TDO), Federated Learning (FL), and Deep Convolutional Neural Network (Deep CNN) to provide a higher accuracy of about 100% and F1-Score of about 99.90%. Furthermore, blockchain integration improves transparency, protects the details of transactions, and safeguards from fraud actions using the SCF system. This research addresses progressing SCF optimization by integrating AI and Blockchain, providing a climbable, well-organized, and protected resolution for real-world entities.
AI's merger with Web3 tech is changing the game leading to open, see-through, and somewhat self-running systems. This research explores the AI-agent paradigm in Web3 considering spread-out networks, blockchain rules, smart deals, and partial self-rule as new concepts. The main focus of this paper is to redefine the integration of artificial intelligence (AI) with Web3 technologies and create a semi-autonomous architecture that transcends decentralized and centralized approaches. While the majority of literature surveys AI agents that function on blockchain and decentralized protocols, our study presents a layered model that exploits off-chain AI inference with on-chain consensus mechanisms, (DID) management, and governance. We explain the main components, such as shared record-keeping, distributed ID management, reward systems, and agreement methods, that allow AI agents to function efficiently in the absence of a central boss. Moreover, the research analyzes significant issues such as scalability, safety, data privacy, and interoperability. It offers a number of improvements in off-chain AI-based agents for decentralized environments. Results demonstrate that hybrid on-chain/off-chain AI clusters can reduce inference costs and increase transaction throughput while preserving decentralization and data privacy. The team-up of AI and Web3 opens doors to new uses like spread-out money systems (DeFi), self-running groups (DAOs), and marketplaces without middlemen creating tough, clear, and user-focused digital worlds.
Filippo Scaramuzza, Renato Cordeiro Ferreira, Giovanni Quattrocchi, Damian Andrew Tamburri · 5 authors
Classical software verification and validation techniques, such as procedural audits, formal methods, or model documentation, are the traditional mechanisms used to achieve the verifiable accountability now required by regulations like the EU AI Act. These methods are either expensive or heavily manual, and ill-suited for the opaque, "black box" nature of most Artificial Intelligence (AI) models. A conflict arises: high auditability and verifiability are required by law, but such transparency conflicts with the need to protect the assets being audited (e.g., confidential data and proprietary models). This paper introduces ZKMLOps, an \ac{MLOps} verification framework that operationalizes Zero-Knowledge Proofs (ZKPs) within Machine-Learning Operations lifecycles; a ZKP allows a prover to convince a verifier that a statement is true without revealing any information about the statement itself. By integrating ZKP with established software engineering patterns, ZKMLOps provides a modular and repeatable process for generating verifiable cryptographic evidence-proofs of well-defined computational statements about the audited model and its inputs-that auditors can use as input to a regulatory compliance determination. We evaluate the framework along two dimensions. First, framework viability: orchestration overhead is bounded and stable across architecturally heterogeneous ZKP backends and models of increasing size. Second, cost-versus-assurance trade-offs: the audit-on-demand setting is the regime in which full zero-knowledge auditing is the appropriate tool, where it provides confidentiality and integrity guarantees that lighter-weight alternatives cannot match.
Financial forecasting increasingly uses large neural network models, but their opacity raises challenges for trust and regulatory compliance. We present several approaches to explainable and reliable AI in finance. \emph{First}, we describe how Time-LLM, a time series foundation model, uses a prompt to avoid a wrong directional forecast. \emph{Second}, we show that combining foundation models for time series forecasting with a reliability estimator can filter our unreliable predictions. \emph{Third}, we argue for symbolic reasoning encoding domain rules for transparent justification. These approaches shift emphasize executing only forecasts that are both reliable and explainable. Experiments on equity and cryptocurrency data show that the architecture reduces false positives and supports selective execution. By integrating predictive performance with reliability estimation and rule-based reasoning, our framework advances transparent and auditable financial AI systems.
Watermarking schemes for large language models (LLMs) have been proposed to identify the source of the generated text, mitigating the potential threats emerged from model theft. However, current watermarking solutions hardly resolve the trust issue: the non-public watermark detection cannot prove itself faithfully conducting the detection. We observe that it is attributed to the secret key mostly used in the watermark detection -- it cannot be public, or the adversary may launch removal attacks provided the key; nor can it be private, or the watermarking detection is opaque to the public. To resolve the dilemma, we propose PVMark, a plugin based on zero-knowledge proof (ZKP), enabling the watermark detection process to be publicly verifiable by third parties without disclosing any secret key. PVMark hinges upon the proof of `correct execution' of watermark detection on which a set of ZKP constraints are built, including mapping, random number generation, comparison, and summation. We implement multiple variants of PVMark in Python, Rust and Circom, covering combinations of three watermarking schemes, three hash functions, and four ZKP protocols, to show our approach effectively works under a variety of circumstances. By experimental results, PVMark efficiently enables public verifiability on the state-of-the-art LLM watermarking schemes yet without compromising the watermarking performance, promising to be deployed in practice.
Blockchains face inherent limitations when communicating outside their own ecosystem, largely due to the Byzantine Fault Tolerant (BFT) 3f+1 security model. Trusted Execution Environments (TEEs) are a promising mitigation because they allow a single trusted broker to interface securely with external systems. This paper develops a cost-of-collusion principal-agent model for compromising a TEE in a Data Center Execution Assurance design. The model isolates the main drivers of attack profitability: a K-of-n coordination threshold, independent detection risk q, heterogeneous per-member sanctions F_i, and a short-window flow prize (omega) proportional to the value secured (beta times V). We derive closed-form deterrence thresholds and a conservative design bound (V_safe) that make collusion unprofitable under transparent parameter choices. Calibrations based on time-advantaged arbitrage indicate that plausible TEE parameters can protect on the order of one trillion dollars in value.
Prajakta Sudhir Khade, Aarushi Santosh Gode, Rajeshkumar U. Sambhe
The exponential rise of cyber threats has revealed the vulnerabilities of centralized security systems, including susceptibility to insider attacks, single points of failure, and regulatory inefficiencies. This paper investigates blockchain as a transformative backbone for cybersecurity, focusing on its potential to ensure data integrity, decentralize trust, and mitigate advanced cyber risks. Beginning with a comprehensive literature review, the study examines the fundamentals of blockchain technology—distributed ledgers, consensus mechanisms, and cryptographic primitives—that enable tamper-proof, transparent, and secure digital ecosystems. The challenges of centralized systems are contrasted with blockchain’s resilience, highlighting its role in eliminating bottlenecks and enhancing trust. Applications across identity management, IoT security, supply chains, and e-governance are analyzed alongside a proposed methodology that integrates blockchain with artificial intelligence, IoT, and quantum-resilient models. Real-world case studies demonstrate blockchain’s adoption in healthcare, government, and industrial systems, while challenges such as scalability, interoperability, and compliance are critically assessed. Collectively, this study underscores blockchain’s pivotal role in shaping next-generation cybersecurity architectures.
Arfa Mahvish, V Surekha, K Archana, Vaseem Ahmed Qureshi · 6 authors
The fast spurts of Internet of Things (IoT) devices have brought new challenges in security and privacy, such as data breaches, unauthorized access, and vulnerability of centralized systems. Traditional IoT security models are based on the centralized architectures and, therefore, have a lack of redundancy and a vulnerability to cyber-attacks. This paper introduces a secured decentralized IoT network based on blockchain technology to improve trust, integrity of data, and security in IoT ecosystem. The distributed ledger technology (DLT) of blockchain makes data transactions tamper-proof, access control transparent and based on decentralized authentication. By integrating smart contracts, the system automates secure device-to-device communication without intermediaries, reducing latency and improving efficiency. The proposed approach eliminates data silos, enhances network resilience, and mitigates security threats such as data manipulation, DDoS attacks, and unauthorized access. Performance analysis demonstrates that blockchain-based IoT security significantly improves data integrity, transaction transparency, and system reliability, making it an ideal solution for smart cities, healthcare, and industrial IoT applications.
Distributed Ledger Technologies (DLTs) underpin Digital Circular Economy (DCE) systems that rely on efficient IoT data flows. Shimmer, a DAG-based DLT optimized for IoT, enables feeless transactions with parallel validation through its tip-selection mechanism. On such ledgers, message fragmentation induces a latency–throughput tradeoff as per-block cost rises with parallel validation. Such efficiency lowers energy and congestion, supporting DCE objectives. Yet, end-users cannot control payload size or network load, leading to unpredictable latency and high CPU use on submitting devices, increasing energy consumption. Existing approaches mostly modify ledger internals, overlooking adaptivity or end-user policies. We introduce ABS-TD3, an offline-to-online TD3 agent that receives the total message size and outputs the optimal per-block size for balancing latency and energy-efficient CPU utilization. The agent is pre-trained offline on real data with Retrieval Augmentation and adaptive weights for improved decision making, then transitioned online with prioritized replay and a novelty bonus, balancing exploitation-exploration, yielding stable adaptivity compared to standard RL approaches. ABS-TD3 is implemented on Shimmer and can integrate with future Tangle-based forks of pre-IOTA-Rebased frameworks, exposing the same client-side controls. ABS-TD3 is evaluated on Shimmer by submitting 8 message sizes ranging from 5KB to 100KB, under the 32 KB block-size limit, with 250 iterations per size via IOTA-SDK. Against max, min, random, and fixed-weight baselines, it reduces median latency by about 9 % to 12 % and median CPU utilization by about 12 % to 17 % versus max and random policies, enabling efficient IoT data submission for DCE platforms without altering DLT infrastructure.
S K Sharif, C H Saritha, P. Senthil, Madhavi Pingili · 6 authors
Every business operation worldwide adopts cloud storage solutions since cybersecurity now demands mandatory protection for data security together with integrity management while also ensuring data confidentiality. Cloud storage systems that run from one central platform remain exposed to cyberattacks that lead to two risks: system malfunctions and unapproved system access. This research delivers an unalterable data management system through the application of blockchain-based methods to distributed cloud architectures. Through the combination of smart contracts with distributed ledger technology (DLT) and cryptographic hashing capabilities in blockchain technology data protection and data integrity get enhanced in cloud systems. Research teams develop hybrid blockchain systems by combining several systems using external storage methods to solve scalability issues. Through shading technology integration with hybrid blockchain systems and off-chain storage systems fast transaction execution becomes possible. The setup of distributed control centers employing blockchain technology secures data better because it extends traditional systems by creating comprehensive visibility that detects unauthorized access attempts. Researchers have investigated how blockchain-enabled cloud storage applications protect digital data in this study.
Saurabh V. Magdum, Sonali Patil, Deepali Nilesh Naik
Federated Learning (FL) revolutionized the field preserving machine learning by facilitating collaborative model training among decentralized clients in absence of raw data. The classic architectures of FT, in contrast, usually rely on a centralized aggregator, which poses threats such as single points of failure, data poisoning, and model inversion attacks. Use of combination of Blockchain technology holds the promise solution via replacement of centralized aggregators with decentralized consensus mechanisms, improving trust, transparency, and data integrity. The present bibliometric analysis considers the correlation of Blockchain and Federated Learning (BFL), with special reference on flagship aggregation algorithms like FedAvg, FedProx, and FedBN, specifically the blockchain networks such as Ethereum, Hyper- ledger Fabric, and Polkadot. Additionally, the paper records actual- world use cases in privacy-sensitive applications like healthcare, finance, and IoT, using benchmark datasets such as MIMIC-III, NASDAQ stock data, and EdgeIIoTset. The proposed study identifies Key trends, timeless findings, and future directions In BFL, gaining perceptual insights of its growing significance for building trustworthy, privacypreserving AI systems.
We present a declarative approach to on-chain limit order books (LOBs) that prioritizes formal verification over raw throughput. Unlike automated market makers, LOBs offer granular control and capital efficiency but are difficult to verify when implemented imperatively in Solidity. Using Pint, a declarative domain-specific language, we encode LOB matching logic, price-time priority, partial fills, and asset conservation, as first-order constraints. Off-chain solvers compute valid state transitions, while the blockchain performs lightweight constraint verification. We implement eight LOB predicates and evaluate performance using real-world transaction traces. Our declarative LOBs achieve 141 predicates/s for simple operations and 11 predicates/s for complex settlement with 1,000 accounts. Performance correlates strongly with state access patterns rather than constraint complexity. Critically, our approach eliminates verification challenges that make imperative smart contracts hard to formally verify, such as unbounded loops, recursion, cross-contract/function calls, and complex control flow. This enables correctness-by-construction through constraint satisfaction, removing the need to prove implementation conformance to specifications. This work demonstrates the first practical evidence that declarative LOBs achieve reasonable performance while providing superior verification guarantees for DeFi protocols.
This paper presents the first rigorous empirical investigation into a fundamental question of cryptocurrency valuation: Are cryptocurrency prices in line with the prices of fundamental assets? To answer this, we analyze the nine largest cryptocurrencies by market capitalization—Bitcoin (BTC), Ethereum (ETH), Solana (SOL), Binance Coin (BNB), Ripple (XRP), Cardano (ADA), Litecoin (LTC), Tron (TRX), and the stablecoin DAI—against a suite of traditional benchmarks, including major fiat currencies (EUR, CAD, JPY), gold, and the S&P500 index. Our dataset spans from 1 January 2014 to 30 June 2025, with start dates varying for newer cryptocurrencies to ensure robust time series analysis. Guided by the asset pricing theory, we formulate a martingale test: if a cryptocurrency is priced in line with a fundamental numeraire asset, its price ratio relative to that numeraire must follow a martingale process. Our extensive empirical analysis reveals that the prices of major cryptocurrencies (BTC, ETH, SOL, BNB) consistently reject the martingale hypothesis when traditional assets (currencies, gold, equities) serve as the numeraire, indicating a decoupling from fundamental valuation anchors. Conversely, when Bitcoin or Ethereum itself is used as the numeraire, most smaller cryptocurrencies are priced in line with these crypto benchmarks, suggesting an internal valuation ecosystem that operates independently of traditional finance.
Oct 30, 2025·2025 1st IEEE Uttar Pradesh Section Women in Engineering International Conference on Electrical Electronics and Computer Engineering (UPWIECON)
The decentralized structure, lack of regulation, and susceptibility to manipulation of Bitcoin markets result in a high level of volatility, which presents substantial obstacles to the accurate prediction of prices. Traditional statistical models, like ARIMA, frequently fall short in describing the dynamic and nonlinear nature of bitcoin markets. In order to overcome this constraint, this research utilizes sophisticated machine learning and deep learning techniques, including as Convolutional Neural Networks (CNN), Decision Trees, Long Short-Term Memory (LSTM), and Logistic Regression, to predict changes in the price of Bitcoin. Using historical Bitcoin datasets, the suggested models are trained and assessed using performance measures like accuracy and RMSE. In comparison to traditional techniques, experimental results show that deep learning models—in particular, LSTM—achieve greater prediction accuracy, offering a more dependable framework for forecasting bitcoin prices.
The rapid proliferation of Internet of Things (IoT) devices across critical sectors has introduced unprecedented security challenges. While Artificial Intelligence (AI) provides powerful tools for threat detection, AI models themselves are vulnerable to data poisoning and adversarial attacks. This paper introduces a novel hybrid framework that synergistically integrates AI with blockchain technology to create a secure and resilient data transmission ecosystem for the IoT. Our architecture utilizes AI-driven techniques, including simulated federated learning, for privacy-preserving anomaly detection at the network edge. A blockchain-based ledger then provides tamper-resistant data validation and immutable record-keeping. The system employs smart contracts on a private Ethereum testnet, combined with the InterPlanetary File System (IPFS) for efficient off-chain storage, ensuring both data integrity and provenance. We evaluated the framework in a simulated environment using benchmark datasets, including CICIDS2017 and UNSW-NB15. The results show a high efficacy in detecting sophisticated attacks, with a Random Forest classifier achieving 98.46
Securing data in the Industrial Internet of Things (IIoT) is critical due to the growing complexity and scale of industrial networks. Integrating Ethereum-based blockchain technology offers a promising solution by leveraging distributed ledgers to enhance the transparency, immutability, and security of IIoT systems. This study aims to enhance threat detection and data protection in IIoT environments through blockchain integration. To achieve this, the proposed approach incorporates Dynamic Threat Landscape (DTL)-based Intrusion Detection Systems (IDS) for real-time attack modelling, enabling systems to adapt to evolving threats. However, integrating blockchain with IIoT also presents challenges, including ensuring low latency for real-time processing, scalability to manage large volumes of sensor data, and maintaining robust cybersecurity while preserving data privacy and integrity. Addressing these concerns is essential for the effective deployment of blockchain-enabled threat detection in IIoT networks.
Short-shelf-life foods have received increasing attention in daily dietary practices due to their freshness and convenience. Compared with other food categories, they impose more stringent requirements on quality assurance and end-to-end traceability. To address the challenge of balancing timeliness in high-frequency real-time data collection with the authenticity of trusted information flow in short-shelf-life food supply chains, this paper proposes a trusted information collection mechanism based on lightweight blockchain technology. First, a multi-layer collaborative architecture is designed, encompassing the perception layer, edge gateway, blockchain layer, and application layer. By integrating off-chain storage with on-chain indexing, the mechanism effectively alleviates the storage and computational burden on the main blockchain. Second, the edge gateway layer incorporates a zero-knowledge interval proof module, enabling real-time, localized privacy compliance statements for sensitive data. Meanwhile, the blockchain layer adopts an PoA consensus mechanism, whereby authorized nodes perform rapid data verification and deposition. Finally, through theoretical analysis and simulation-based validation, the results demonstrate that the proposed mechanism not only enhances the real-time performance, authenticity, and privacy protection of data in short-shelf-life food supply chains, but also achieves high operational efficiency and scalability. This work thus provides a novel theoretical foundation and practical approach for trusted information collection in the context of short-shelf-life food supply chains.