Identifying Tor Anonymous Traffic Based on Gravitational Clustering Analysis
Ziye Deng, Guoqing Qian, Zhouguo Chen, Hong Su · 2017
The anonymous communication technology has brought new challenges to network monitoring. Effectively identify the anonymous traffic, plays a key role in preventing the abuse of such technology. In this paper, we propose a gravitational clustering algorithm (GCA) to identify the Tor anonymous traffic. Basically, each vector in the dataset is considered as an object in the feature space. And the objects are moved by using gravitational force and the second motion law. Compared with traditional method, our method could automatically identify the cluster number. And Clustering algorithm could help us discover the anonymous traffic among various unknown traffic types. We also make an empirical comparison of current state-of-the-art clustering algorithms. Experimental results show that our method has a better performance compared with other methods under the same experimental settings.