AToM: Active Topology Monitoring for the Bitcoin Peer-to-Peer Network
This addresses the need for better network analysis in cryptocurrencies, though it is incremental as it builds on existing inference methods.
The paper tackles the problem of hidden connections in the Bitcoin P2P network, which hinders analysis for efficiency and security, by proposing a protocol to infer and monitor these connections with over 90% precision and recall even with up to 20% malicious nodes.
Over the past decade, the Bitcoin P2P network protocol has become a reference model for all modern cryptocurrencies. While nodes in this network are known, the connections among them are kept hidden, as it is commonly believed that this helps protect from deanonymization and low-level attacks. However, adversaries can bypass this limitation by inferring connections through side channels. At the same time, the lack of topology information hinders the analysis of the network, which is essential to improve efficiency and security. In this paper, we thoroughly review network-level attacks and empirically show that topology obfuscation is not an effective countermeasure. We then argue that the benefits of an open topology potentially outweigh its risks, and propose a protocol to reliably infer and monitor connections among reachable nodes of the Bitcoin network. We formally analyze our protocol and experimentally evaluate its accuracy in both trusted and untrusted settings. Results show our system has a low impact on the network, and has precision and recall are over 90% with up to 20% of malicious nodes in the network.