A Real-Time Anonymous Traffic Detection Based on Reinforcement Learning

Abstract

Anonymous networks have been utilized to protect user anonymity. However, such networks have been a platform for network attacks. Therefore, detecting anonymous network traffic is critical for defending a network. Many machine learning and deep learning-based approaches have been proposed. However, many of them rely heavily on labeled data and have intricate structures, which may degrade their real-time performance. Aiming to mitigate these issues, this study proposes a lightweight system to detect real-time anonymous network traffic by harnessing the principles of reinforcement learning. Initially, the historical traces of anonymous traffic are analyzed to identify the attributes that characterize anonymous and regular network traffic. Building on these important attributes, we design three components: states, actions, and rewards. Operating autonomously, the system employs these elements to discern network traffic categories in an unsupervised mode. Empirical results demonstrate that the system can identify patterns in anonymous traffic with an accuracy rate surpassing 80%.