Transaction Deanonymization in Large-Scale Bitcoin Systems via Propagation Pattern Analysis
Bitcoin is a digital currency payment system, which bases on the property of decentralization and anonymization of Blockchain. Researches on transaction deanonymization for the Bitcoin system may not associate anonymous transactions with the IP addresses
- PDF / 1,900,567 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 67 Downloads / 169 Views
School of Computer Science, Beijing Institute of Technology, Beijing 100081, China {shenmeng,duanjx,liehuangz}@bit.edu.cn, [email protected] 2 Cyberspace Security Research Center, Peng Cheng Laboratory, Shenzhen 518000, China
Abstract. Bitcoin is a digital currency payment system, which bases on the property of decentralization and anonymization of Blockchain. Researches on transaction deanonymization for the Bitcoin system may not associate anonymous transactions with the IP addresses (physical identity) of the originator accurately and may consume network resources excessively. In this paper, we propose an approach to obtain the originating transactions through analyzing the propagation information. We calculate a pattern matching score by combining the propagation pattern extraction and the node weight assignment. Through carrying out the experiments in the real Bitcoin system, we effectively match the originating transactions with the target node, which reaches a precision of 81.3% and is 30% higher than the state-of-the-art method. Keywords: Bitcoin transactions · Deanonymization · Data analytics Propagation path · Empirical probability distribution
1
·
Introduction
Bitcoin system has been recently emerging as a world-wide decentralized anonymous digital currency system [1]. It works without a centralized administrator, where all transactions are verified by nodes in a peer-to-peer (P2P) Bitcoin network and recorded in public distributed ledger (i.e., Blockchain) [7,15–17]. The total number of nodes (a.k.a., peers) in the Bitcoin P2P network is estimated to exceed 10,300. At the same time, the system generates a block every ten minutes and 2,000 transactions per block on average. In the complex system, assuming to track who issued a specific transaction, we need to match the transactions with the nodes [19,23]. However, matching large-scale transactions with their This work is partially supported by Key-Area Research and Development Program of Guangdong Province (No. 2019B010137003), Zhejiang Lab Open Fund with No. 2020AA3AB04, National Natural Science Foundation of China under Grants 61972039 and 61872041, and Beijing Natural Science Foundation under Grant 4192050. c Springer Nature Singapore Pte Ltd. 2020 S. Yu et al. (Eds.): SPDE 2020, CCIS 1268, pp. 661–675, 2020. https://doi.org/10.1007/978-981-15-9129-7_45
662
M. Shen et al.
originating nodes needs a massive information collection and extraction, which may lead to prohibitively expensive computational and storage costs [20–22]. We aim to design an effective and efficient method to identify the Bitcoin transaction of originators (i.e., peers who generate transactions) with its IP addresses. Existing studies on Bitcoin transaction deanonymization can be roughly classified into two categories. The first category mainly focuses on discovering the relationship among Bitcoin anonymous peers from publicly available transactions, e.g., extracting user relationships through transaction graph analysis [5,11,18], or clustering transaction addresses belonging
Data Loading...
