| 摘要: |
| 针对以太坊网络拓扑隐藏的特征以及当前拓扑感知方法精度较差的问题, 提出了一种基于通信协议和未来事务特征的以太坊活跃网络拓扑感知方法。首先, 通过改进的以太坊节点发现协议构建虚拟地址并主动对目标节点路由表进行映射, 感知目标节点的潜在活跃连接节点集合。其次, 向以太坊网络中部署感知节点, 通过收集目标节点转发的消息本体与消息哈希的数量,并根据两者之间的比例关系, 对目标节点的活跃连接数量进行推测。 最后, 根据以太坊内存池中未来事务的隔离特征, 本文提出了一种基于隔离验证的节点间活跃连接验证的改进方法。通过面向以太坊测试网络展开广泛的实验, 本文所提出的方法能够在理想的时间内感知以太坊的区块链覆盖层网络拓扑快照, 相较于现有方法所感知的网络拓扑具有更高的精确性。根据实验数据表明, 以太坊网络中活跃节点不足整个网络的2%, 而且活跃节点出现聚集和重叠现象。通过对以太坊网络拓扑进行建模, 分析可得以太坊网络具有较小的平均最短路径, 信息传播在网络中表现出小世界效应, 并且节点的度值遵循幂律分布, 符合以太坊分布式网络的设计初衷。 |
| 关键词: 以太坊网络 拓扑感知 网络分析 点对点网络 |
| DOI：10.19363/J.cnki.cn10-1380/tn.2024.07.06 |
| 投稿时间：2022-10-24修订日期：2023-02-20 |
| 基金项目:本课题得到江苏省研究生科研与实践创新项目(No. SJCX21_0115)资助。 |
|
| Research on Active Network Topology Perception and Analysis for Ethereum |
| BAI Xiaodong,LIU Daidong,WEI Songjie |
| School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China |
| Abstract: |
| This paper proposes a topology perception method for the Ethereum active network that is based on the Ethereum communication protocol and future transaction characteristics. The purpose of this method is to uncover the hidden features of the Ethereum network topology and improve the accuracy of the methods that are currently used to perceive the topology of networks. Firstly, the set of potentially active connected nodes of the target node are determined by creating virtual addresses and actively mapping them to the target node's routing table using an enhanced Ethereum node discovery protocol. Secondly, deploy sensing nodes to the Ethereum network, gather the number of message ontology and message hashes delivered by the target node, and speculate on the number of active connections of the target node based on their proportionate relationship. Finally, based on the isolation features of future transactions in the Ethereum memory pool, this article presents an enhanced mechanism for Isolated Witness-based active connection verification across nodes. Extensive experiments on the Ethereum test network have demonstrated that the approach presented in this research can perceive the network topology snapshot of Ethereum's blockchain overlay layer in an optimal amount of time and with greater accuracy than existing methods. precision. According to experimental data, less than 2% of the nodes in the Ethereum network are active, and the active nodes appear to cluster and overlap. Through modeling of Ethereum network topology, the analysis demonstrates that the Ethereum network has a short average shortest path, that information transmission in the network exhibits a small-world effect, and that the degree value of nodes follows a power law distribution, which is consistent with Ethereum's original design intent. |
| Key words: ethereum network topology perception network analysis peer-to-peer networ |
