摘要
比特币技术发展迅速,交易规模逐步攀升,引起国内外广泛关注.比特币技术具备去中心化和匿名化特征,使得比特币交易者的身份难以识别,为不法行为(如毒品交易、比特币勒索病毒等)提供了隐匿空间.该文提出了一种针对比特币交易的溯源机制,能够追踪交易信息在网络层的传播路径,从而将交易中的匿名比特币地址和发起交易节点的IP地址相关联.通过设计一种基于主动嗅探的邻居节点识别方法,溯源机制支持轻量级监测,而且相比现有溯源技术具有更好的实用性.文中针对比特币系统开发了溯源程序,从有效性、准确率、适用范围等方面对其进行测试与分析评估.实验结果表明,比特币网络中有69.9%的服务器节点适用于这种溯源机制,能够获得召回率50%、准确率31.25%的溯源精度,优于现有的交易溯源方法,具有较强的实践意义和使用价值.
The rapid development of Bitcoin technology and the growing scale of Bitcoin transactions have drawn wide attention at home and abroad.Whereas,Bitcoin is often used by terrorists and criminals attracted to the anonymity of the currency,such as all deals on Silk Road were made in Bitcoin.Therefore,it is essential to supervise Bitcoin and track the source transaction when necessary.However,as Bitcoin technology has the characteristics of de-centralization,traditional financial supervision means cannot provide effective supervision.Philip Koshy et al.found some special trading patterns for originating node by analyzing the propagation law of currency transactions in the network layer,but the proportion of special deals is less than 9%.Alex Biryukov et al.take advantages of the information of neighbor nodes of Bitcoin peer to locate the originating node.This approach improves fault tolerance and accuracy(experiment shows the accuracy of 11%),but requires constantly sending information to all nodes,which can cause network congestion.There are also some methods of transaction data analysis.However,they usually only get the relationship between the addresses,but cannot directly obtain the corresponding identity information of the address.Therefore,it is necessary to design a new transaction tracking mechanism for Bitcoin architecture,which can detect with fewer resources and has higher tracking accuracy than existing mechanisms.In this paper,we optimize the existing bitcoin transaction traceability mechanism and propose a new neighbor node identification scheme based on active sniffing.Our scheme supports lightweight transaction traceability and has better traceability than existing schemes.In addition,by designing a matching value optimization method based on multiple detection,the traceability mechanism can gradually improve the traceability results through continuous monitoring and improve the traceability accuracy.The main contributions of this paper include three parts:at first,we designs a practical Bitcoin transaction tracking mechanism that can track the transmission of bitcoin transactions under the public Bitcoin network and associate the anonymous bitcoin transaction with the IP address of the transaction originating node.Secondly,for the first time,we propose a new method for neighbor node detection based on active sniffing,which can infer the neighbor nodes of a specific node by sending probe information.This method can obtain the topology information of any server node with less resources.Finally,we developed a prototype system for traceability mechanisms and tested the efficiency and accuracy on public Bitcoin network.The experiment results demonstrate that 69.9%of the backbone nodes in the Bitcoin network are suitable for the proposed tracing mechanism,with traceability recall rate of 50%and accuracy of 31.25%,which is superior to the current tracing methods and of great importance in practice.The proposed traceability mechanism can trace the transactions in Bitcoin networks and identify the transactions created by specific server nodes which can help to track down criminals who maliciously use bitcoin technology to deter Bitcoin-based crimes.Moreover,the traceability mechanism of this article is also applicable to altcoin based on Bitcoin code and other digital currencies based on Blockchain technology,and has a wide range of application scenarios.
作者
高峰
毛洪亮
吴震
沈蒙
祝烈煌
李艳东
GAO Feng;MAO Hong-Liang;WU Zhen;SHEN Meng;ZHU Lie-Huang;LI Yan-Dong(School of Computer Science,Beijing Institute of Technology,Beijing 100081;National Computer Network Emergency Response Technical Team/Coordination Center of China(CNCERT/CC),Beijing 100029)
出处
《计算机学报》
EI
CSCD
北大核心
2018年第5期989-1004,共16页
Chinese Journal of Computers
基金
本课题得到国家重点研发计划(2016YFB0800301)、国家自然科学基金(61602039)、北京市自然科学基金(4164098)、广西云计算与大数据协同创新中心(YD16E14)、CCF-启明星辰科研基金项目资助.
关键词
比特币
对等网络
溯源
区块链
反匿名
Bitcoin
peer-to-peer networking
tracing
blockchain
de-anonymization
