Anonymous web browsing through predicted pages

Shui Yu,Theerasak Thapngam,Hou I. Tse,Jilong Wang
DOI: https://doi.org/10.1109/GLOCOMW.2010.5700205
2010-01-01
Abstract:Anonymous web browsing is an emerging hot topic with many potential applications for privacy and security. However, research on low latency anonymous communication, such as web browsing, is quite limited; one reason is the intolerable delay caused by the current dominant dummy packet padding strategy, as a result, it is hard to satisfy perfect anonymity and limited delay at the same time for web browsing. In this paper, we extend our previous proposal on using prefetched web pages as cover traffic to obtain perfect anonymity for anonymous web browsing, we further explore different aspects in this direction. Based on Shannon's perfect secrecy theory, we formally established a mathematical model for the problem, and defined a metric to measure the cost of achieving perfect anonymity. The experiments on a real world data set demonstrated that the proposed strategy can reduce delay more than ten times compared to the dummy packet padding methods, which confirmed the vast potentials of the proposed strategy.
What problem does this paper attempt to address?