Financial services must monitor their transactions to prevent being used for money laundering and combat the financing of terrorism. Initially, organizations in charge of fraud regulation were only concerned about financial institutions such as banks. However, nowadays, the Fintech industry, online businesses, or platforms involving virtual assets can also be affected by similar criminal schemes. Regardless of the differences between the entities mentioned above, malicious activities affecting them share many common patterns. This dissertation's first goal is to compile and compare existing studies involving machine learning to detect and analyze suspicious transactions. The second goal is to synthesize methodologies from the last goal for tackling different use cases in an organized manner. Finally, the third goal is to assess the applicability of deep generative models for enhancing existing solutions. In the first part of the thesis, we propose an unsupervised methodology for detecting suspicious transactions applied to two case studies. One is related to transactions from a money remittance network, and the other is related to a novel payment network based on distributed ledger technologies. Anomaly detection algorithms are applied to rank user accounts based on recency, frequency, and monetary features. The results are manually validated by domain experts, confirming known scenarios and finding unexpected new cases. In the second part, we carry out an analogous analysis employing supervised methods, along with a case study where we classify Ethereum smart contracts into honeypots and non-honeypots. We take features from the source code, the transaction data, and the funds' flow characterization. The proposed classification models proved to generalize well to unseen honeypot instances and techniques and allowed us to characterize previously unknown techniques. In the third part, we analyze the challenges that tabular data brings into the domain of deep generative models, a particular type of data used to represent financial transactions in the previous two parts. We propose a new model architecture by adapting state-of-the-art methods to output multiple variables from mixed types distributions. Additionally, we extend the evaluation metrics used in the literature to the multi-output setting, and we show empirically that our approach outperforms the existing methods. Finally, in the last part, we extend the work from the third part by applying the presented models to enhance classification tasks from the second part, commonly containing a severe class imbalance. We introduce the multi-input architecture to expand models alongside our previously proposed multi-output architecture. We compare three techniques to sample from deep generative models defining a transparent and fair large-scale experimental protocol and interesting visual analysis tools. We showed that general machine learning detection and visualization techniques could help address the fraud detection domain's many challenges. In particular, deep generative models can add value to the classification task given the imbalanced nature of the fraudulent class, in exchange for implementation and time complexity. Future and promising applications for deep generative models include missing data imputation and sharing synthetic data or data generators preserving privacy constraints.
Energy supply industries play a vital role in a country. Inefficiencies in the energy supply chain regarding tricky contest and the lack of management instantly change energy tariff calculation. This work proposes the Ethereum blockchain platform with existing traditional infrastructure to track and investigate energy supply chain activities using a unique identity with smart contracts. It maintains the records of the organization's activities that are protected and available to stakeholders according to the recognized collection of procedures and practices without requiring any centralized administration. This paper focuses entirely on analyzing and developing a simplified, low-cost economical solution to quickly connect the present energy supply industry at various geological locations to track and trace the linked data in the energy market.
It is widely accepted that Ethereum mining is highly centralized.\nNonetheless, centralization has been mostly characterized by exclusively\nlooking at the influence that independent miners or mining pools can have over\nthe network. Moreover, models of mining behavior assume that miners are either\nunrelated or only relate via mining pools under highly structured and\ntransparent agreements. If these assumptions and the predictions they entail\nwere to be completely accurate, there would not be any evidence of on-chain\ntransactions between miners, other than the ones expected from mining pool\npayouts. By looking at on-chain transactions between miners in the Ethereum\nNetwork we find that aside from the payouts from mining pools to small miners,\nthere are also transactions that define relationships between mining pools,\nindependent miners and between independent miners and mining pools.\nFurthermore, by characterizing the topology of the network of miner\ntransactions, we find the emergence of highly connected clusters that control\nsignificant amounts of hashing power and exhibit relationships in the opposite\ndirection of what theoretical models predict. This more nuanced\ncharacterization of mining centralization can help identify network\nvulnerabilities and inform protocol redesigns.\n
Abstract This study examines the informational efficiency of the bitcoin spot market by evaluating the predictive power of mechanical trading rules designed to exploit price continuation. Significant return predictability is found until the introduction of bitcoin futures in December 2017. The forecasting ability of trend‐chasing trading rules declines dramatically afterwards. Although evidence suggests that the introduction of bitcoin futures has increased the informational efficiency of the bitcoin spot market, no signs of improvement in informational efficiency are found in ethereum, the second‐largest cryptocurrency—following the introduction of bitcoin futures.
This research is dedicated to the development of a prototype of open infrastructure for users' internet traffic filtering on a browser level. We described the advantages of a distributed approach in comparison with current centralized solutions. Besides, we suggested a solution to define the optimum size for a URL storage block in Ethereum network. This solution may be used for the development of infrastructure of DApps applications on Ethereum network in future. The efficiency of the suggested approach is supported by several experiments.
Using hard-fork mechanism on the blockchain to recover the losses caused by attacks contradicts the immutable characteristic of a blockchain system. To prevent malicious transactions from getting into blockchains in advance, we propose a runtime hook technique to synchronize and analyze the ongoing transactions exposed to the Ethereum transaction pool. Having a complete view of the past and the ongoing transactions, we can identify and enforce abortion of malicious transactions and prevent losses due to attacks being executed and recorded in the blockchain. Specifically, we modify the Ethereum source code to instrument the entry point of a node to synchronize data received from the Ethereum P2P network and systematically scan suspicious patterns in transactions to identify potential attacks. As a proof-of-the-concept, we show how to deploy the proposed runtime hook system on a private blockchain system, such that we can detect and prevent transactions of double spending on the 51% attack and reentrancy attack of smart contracts.
Abhilash Kancharla, Zuqiang Ke, Nohpill Park, Hye-Young Kim
The hybrid chain is proposed in this paper to investigate on a new blockchain network that is to be built across private and main nets, namely, a hybrid chain. Note that the hybrid chain is distinguished from the on/off-balanced chain such that the hybrid chain is across two different (e.g., Hypercubes private net vs. Ethereum main net) while the on/off-balanced chain across on-chain and off-chain (e.g., Ethereum main net vs. cloud). It is essential to build a dependable interface in between private and main nets if business to consumer (or vice versa) transactions are demanded for instance. In the course of interfacing across private and main nets, dependability is to be considered as one of the most critical design factors in order to ensure private transactions stay within the private territory and publicized transactions stay public in the main net, and further in order to facilitate a seamless yet dependable execution of transactions across the border. In this context, the efficacy of the privacy of the private net side and the publicity of the main net side will be addressed and modeled by tracing a transaction's stochastic process at a steady state. A protype for an isolated testing across the Ethereum and Hypercubes open source for validation purpose is proposed along with extensive parametric simulations.
This paper proposes a novel embedded Markovian queueing model of the M(1,n)/Mn/1 type in order to establish a theoretical foundation to design a blockchain-based system with specific interest in Ethereum. The model assumes variable bulk arrivals of transactions in Poisson distribution, i.e., M(1,n), where n the number of slots across all the mined transactions and static bulk service of transactions in exponential time, i.e., Mn, for posting in the current block, namely, VBASBP. The primary performance measurements to be taken are the average number of slots no matter how many transactions are mined under the assumption of the maximum number of slots per block as specified by n; the average waiting time per slot; and the throughput in terms of the average number of slots to be processed per time. The variable bulk arrival rate is assumed to vary linearly proportional to the size of the transactions in a multiple of per slot, and the static bulk service is assumed to take place when the number of slots in the mined transactions reaches at n, i.e., a bulk processing of multiple transactions in multiple slots for posting in a block. Numerical simulations are conducted on Matlab to demonstrate the efficacy of the model.
In the medical system, the verification, preservation and synchronization of electronic medical records has always been a difficult problem, and the random dissemination of patient records will bring various risks to patient privacy. Therefore, how to achieve secure data sharing on the basis of ensuring users' personal privacy becomes the key. In recent years, blockchain has been proposed to be a promising solution to achieve data sharing with security and privacy preservation due to its advantages of immutability. So, a distributed electronic medical records searchable scheme was proposed by leveraging blockchain and smart contract technology. Firstly, we perform a hash calculation on the electronic medical data and store the corresponding value on the blockchain to ensure its integrity and authenticity. Then, we encrypt the electronic medical data and store it in the interplanetary file system which is a distributed storage protocol. These operations not only can solve centralized data store of servers of several medical institutions, but also be good at lowering stress from data store and high-frequency access to blockchain. Secondly, the encrypted keyword index information of electronic medical records was stored on the Ethereum blockchain, meanwhile a smart contract deployed in the Ethereum blockchain is used to realize keyword search instead of depending on a centralized third party. Furthermore, we use attribute-based encryption scheme to ensure that only the attributes meeting the access policy can decrypt the encrypted electronic medical records. Finally, our performance analysis and security analysis show that the scheme is secure and efficient.
This paper is devoted to the problems of the short-term forecasting cryptocurrencies time series using machine learning approach. We applied two the most powerful ensembles methods: Random Forests (RF) and Gradient Boosting Machine (GBM). For testing models we used the daily close prices of three the most capitalized coins: Bitcoin (BTC), Ethereum (ETH) and Ripple (XRP), and as a features were selected the past price information and technical indicators (moving average). To check the efficiency of these models we made out-of-sample forecast for three cryptocurrencies by using one step ahead technique. As the accuracy rate for our models we were selected Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) metrics. According to comparative analysis of the predictive ability of the RF and GBM both models showed the same order of accuracy for the out-of-sample dataset prediction, although boosting also was somewhat more accurate. Computer experiments have confirmed the feasibility of using the machine learning ensembles approaches considered for the short-term forecasting of cryptocurrencies time series. Built models and their ensembles can be used as the basis algorithms for automated Internet trading systems.
Ethereum is the typical representative of the second generation of Blockchain technology. To achieve anonymity and decentralization, Ethereum utilizes a structured P2P network for data broadcasting, replication, and synchronization. The Ethereum P2P network topology construction profoundly affects the overall Ethereum system performance and reliability. However, there are few studies on the Ethereum underlying P2P overlay network topology. In this paper, we perform a preliminary investigation on this extensive Ethereum P2P system with a comprehensive measurement method combining active and passive methods. We gather Ethereum nodes information by active method for nodes crawling and verification, and we use the passive network traffic monitoring for routing data gathering. After a period of data collection for more than one month, we get a massive data set that has more than 50 million routing data entries. This data set allows us to gain knowledge on the Ethereum network scale and to derive mathematical characteristics of overlay network structure. By analyzing the data set comprehensively, we find that the Ethereum network exhibits both small-world and scale-free characteristics. There are a few critical nodes that have a significant impact on the robustness and vulnerability of the whole system. The Ethereum network also has a very high concentration in terms of network routing, which is very vulnerable to routing attacks.
The Ethereum cryptocurrency system becomes prevalent with the increasing number of daily transactions. It is a high-quality channel to secretly transfer information because senders and recipients are pseudonymous and transactions are tamper-resistant. Several existing covert channels have been designed for several years. However, most of the existing covert channels can either be detected by their special patterns or have low embedding rate. In this paper, we first use the VALUE field of a transaction in the Ethereum system to construct a HMAC-based multiple-bit embedding (HMAC-based MBE) scheme. We further proposed a Hash-based multiple-bit embedding (Hash-based MBE) scheme to enhance the covertness. Theoretical and experimental analysis show that both schemes have either higher embedding rate or better covertness than that of the existing works.
Blockchain Technology Applications and Security
Advanced Steganography and Watermarking Techniques
Selling digital commodities securely over the Internet is a challenging task when Seller and Buyer do not trust each other. With the advent of cryptocurrencies, one prominent solution for digital exchange is to rely on a smart contract as a trusted arbiter that fairly resolves disputes when Seller and Buyer disagree. Such protocols have an optimistic mode, where the digital exchange between the parties can be completed with only minimal interaction with the smart contract. In this work we present OptiSwap, a new smart contract based fair exchange protocol that significantly improves the optimistic case of smart contract based fair exchange protocols. In particular, OptiSwap has almost no overhead in communication complexity, and improves on the computational overheads of the parties compared to prior solutions. An additional feature of OptiSwap is a protection mechanism against so-called grieving attacks, where an adversary attempts to violate the financial fairness of the protocol by forcing the honest party to pay fees. We analyze OptiSwap's security in the UC model and provide benchmark results over Ethereum.
In this study, a web model which provides price estimation in terms of Turkish Lira for Bitcoin, Ethereum and Ripple which are popular cryptocurrencies, is developed. Using the relevant model, the price estimation of these three crypto currencies between 21.09.2019 and 20.11.2019 is carried out on the web with dynamic data. Artificial intelligence methods such as adaptive neural fuzzy inference system, artificial neural networks, polynomial curve fitting and long short-term memory are used for price estimation. The aim of this study is to provide periodic forecasts to individuals or institutions interested in cryptocurrencies and to test the success of the exemplary model of the use of artificial intelligence in finance. When the forecastings that haven't yet been realized at the relevant dates and the actual values are compared, the successful results show that the model is well established.
The peer-to-peer (P2P) network of blockchain used to transport its transactions and blocks has a high impact on the efficiency and security of the system. The P2P network topologies of popular blockchains such as Bitcoin and Ethereum, therefore, deserve our highest attention. The current Ethereum blockchain explorers (e.g., Etherscan) focus on the tracking of block and transaction records but omit the characterization of the underlying P2P network. This work presents the Ethereum Network Analyzer (Ethna), a tool that probes and analyzes the P2P network of the Ethereum blockchain. Unlike Bitcoin that adopts an unstructured P2P network, Ethereum relies on the Kademlia DHT to manage its P2P network. Therefore, the existing analytical methods for Bitcoin-like P2P networks are not applicable to Ethereum. Ethna implements a novel method that accurately measures the degrees of Ethereum nodes. Furthermore, it incorporates an algorithm that derives the latency metrics of message propagation in the Ethereum P2P network. We ran Ethna on the Ethereum Mainnet and conducted extensive experiments to analyze the topological features of its P2P network. Our analysis shows that the Ethereum P2P network possesses a certain effect of small-world networks, and the degrees of nodes follow a power-law distribution that characterizes scale-free networks.
Smart contracts are Turing-complete programs running on the blockchain. They cannot be modified, even when bugs are detected. The Selfdestruct function is the only way to destroy a contract on the blockchain system and transfer all the Ethers on the contract balance. Thus, many developers use this function to destroy a contract and redeploy a new one when bugs are detected. In this paper, we propose a deep learning-based method to find security issues of Ethereum smart contracts by finding the updated version of a destructed contract. After finding the updated versions, we use open card sorting to find security issues.
Santiago Morales Cardoso, Mario R. Morales, Vilmar Wladimir Chipuxi-Trujillo, Jhonathan Paucar
La presente investigación, tiene como objetivo proponer un modelo de cadena de suministro que sea soportado por la tecnología blockchain a través del desarrollo de contratos inteligentes. Se establecen especificaciones funcionales que cumplan con la propuesta a través de la selección de herramientas tecnológicas, para la construcción de un contrato inteligente, en donde se especifica su distribución como: estructuras, variables de estado y funciones, que brindan solución a los requerimientos del modelo en base a un ejemplo realizado en el sector farmacéutico. Simulando la interacción dentro de una red Ethereum de prueba con la finalidad de analizar el flujo de trazabilidad de un producto, a través de la intervención de participantes dentro de la cadena de suministro como pueden ser: fabricantes, proveedores o clientes.
Case law is the term that refers to reports of past court decisions. It is considered an essential source of law, vital for legal professionals. Existing case law services are currently centralized, with an entity having complete control over the data and often charging fees for its access and other adding value services. This paper attempts to leverage the potential of blockchain technology in order to develop a public and decentralized platform that allows the submission of court decisions in a decentralized database and employs a network of curators who offer their validation, classification, and evaluation. Specifically, we design, analyze and implement AnyCase, a proof-of-concept prototype system on the Ethereum platform. We focus on the establishment of a sybil-resistant voting protocol used for reaching agreement and the development of a tokenized economy that incentivizes participation. Our preliminary analysis indicates that, besides being decentralized, AnyCase has the potential to compete with existing centralized systems in several other aspects.
The article concerns selected issues regarding smart contracts from the perspective of private law, in particular the concept of a contract, determination of its content, principles of performance and breach of an obligation. The legal analysis jest supplemented with technological aspects that show the essence and mechanism of operation of Blockchain, smart contracts, Ethereum. The legal doctrine generally referred to the technological aspects of smart contracts, attempting to include them in the traditional contract law system. The article contends that this is the wrong approach. The authors argue that a smart contract as such is not a contract, but a computer program (code) that can be a manner of concluding a contract and at the same time self-executing it. The qualification of a smart contract as a method of conclusion a contract, and not the contract itself, determines the conclusions on other aspects of concluding the contract, its interpretation, etc. Considerations concerning the model of a self-executing contract are formulated around this thesis. This model determines the proposals for regulation of the principles of performance of the contract, the consequences of the breach of contract and others areas of concern.
Francesco Bruschi, Vincenzo Rana, Alessio Pagani, Donatella Sciuto
The idea of Smart contracts foresees the possibility of automating contractual clauses using hardware and software tools and devices. One of the main perspectives of their implementation is the automation of interactions such as bets, collaterals, prediction markets, insurances. As blockchain platforms, such as Ethereum, offer very strong guarantees of untampered, deterministic execution, that can be exploited as smart contracts substrate, the problem of how to provide reliable information from the “outside world” into the contracts becomes central. In this article, we propose a system based on a Merkle tree representation of structured documents (such as all XML), with which it is possible to generate compact proofs on the content of web documents. The proofs can then be efficiently checked on-chain by a smart contract, to trigger contract action. We provide an end-to-end proof of concept, applying it to real use case scenarios, which allows us to give an estimate of the costs.
It is well known that users on open blockchains are tracked by an industry providing services to governments, law enforcement, secret services, and alike. While most blockchains do not protect their users' privacy and allow external observers to link transactions and addresses, a growing research interest attempts to design add-on privacy solutions to help users regain their privacy on non-private blockchains. In this work, we propose to our knowledge the first censorship resilient mixer, which can reward its users in a privacy-preserving manner for participating in the system. Increasing the anonymity set size, and diversity of users, is, as we believe, an important endeavor to raise a mixer's contributed privacy in practice. The paid-out rewards can take the form of governance tokens to decentralize the voting on system parameters, similar to how popular "DeFi farming" protocols operate. Moreover, by leveraging existing "Defi" lending platforms, AMR is the first mixer design that allows participating clients to earn financial interests on their deposited funds. Our system AMR is autonomous as it does not rely on any external server or third party. The evaluation of our AMR implementation shows that the system supports today on Ethereum anonymity set sizes beyond thousands of users, and a capacity of over $66,000$ deposits per day, at constant system costs. We provide a formal specification of our zksnark-based AMR system, a privacy and security analysis, implementation, and evaluation with both the MiMC and Poseidon hash functions.
Open access
2 source records
cs.CR
Blockchain Technology Applications and Security
Advanced Steganography and Watermarking Techniques