RegulaTor: A Straightforward Website Fingerprinting Defense

Abstract Website Fingerprinting (WF) attacks are used by local passive attackers to determine the destination of encrypted internet traffic comparing sequences packets sent and received user a previously recorded data set. As result, WF particular concern privacy-enhancing technologies such as Tor. In response, variety defenses have been developed, though they tend incur high bandwidth latency overhead or require additional infrastructure, thus making them difficult implement in practice. Some lighter-weight presented well; still, attain only moderate effectiveness against recently published attacks. this paper, we aim present realistic novel defense, RegulaTor, which takes advantage common patterns web browsing reduce both defense accuracy current closed-world setting, RegulaTor reduces state-of-the-art attack, Tik-Tok, comparable from 66% 25.4%. To achieve performance, it requires 6.6% 39.3% less than leading moderate-overhead defense. open-world limits precision-tuned Tik-Tok attack an F 1 -score of. 135, compared .625 for best