Answers are grounded in this paper and cite their sources.
P2P (peer-to-peer) technology is widely used in the internet, so the research on identifying P2P traffic has always been a hot topic. However, in the past, the identification of P2P traffic was only limited to pure IPv4 networks. In the global trend of promoting the deployment of IPv6, universities are in the stage of transition to an IPv6 network environment, where IPv4 and IPv6 will coexist for a long time. Currently, there is a lack of research on the identification of P2P traffic in this complex dual stack environment. To address the limitations of traditional recognition methods that cannot detect encrypted traffic and manually determine popular feature thresholds, this article adopts the C4.5 decision tree algorithm in machine learning, selecting the proportion of upstream and downstream traffic, the degree of discreteness of packet size, and the length of flow connection as sample features, collecting data traffic in a real university dual stack network environment to identify P2P traffic under both IPv4 and IPv6 links. The experimental results show that this method can effectively identify P2P traffic under IPv4 and IPv6, with an average accuracy of over 95%.
No takes yet. Share an insight, caveat, or question.
Xiaodong He (2023) studied this question.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: