New kind of correlation attack: How (ir)relevant is it?

[JSD 6]

Hello,

 

recently I found an interesting article about traffic correlation attacks against anonymity networks. I have much too little knowledge to judge how valid and relevant it actually is, but I would just like to put it up fir discussion:

 

https://arxiv.org/pdf/1808.07285.pdf

 

Specifically, they claim:

 

 

Despite the importance of flow correlation attacks on Tor, existing flow correlation techniques are

considered to be ineffective and unreliable in linking Tor flows when applied at a large scale, i.e., they impose high rates of false positive error rates or require impractically long flow observations to be able to make reliable correlations. In this paper, we show that, unfortunately, flow correlation attacks can be conducted on Tor traffic with drastically higher accuracies than before by leveraging emerging learning mechanisms. We particularly design a system, called DeepCorr, that outperforms the state-of-the-art by significant margins in correlating Tor connections. DeepCorr leverages an advanced deep learning architecture to learn a flow correlation function tailored to Tor’s complex network—this is in contrast to previous works’ use of generic statistical correlation metrics to correlated Tor flows. We show that with moderate learning, DeepCorr can correlate Tor connections (and therefore break its anonymity) with accuracies significantly higher than existing algorithms, and using substantially shorter lengths of flow observations. For in-stance, by collecting only about 900 packets of each target Tor flow (roughly 900KB of Tor data), DeepCorr provides a flow correlation accuracy of 96% compared to 4% by the state-of-the-art system of RAPTOR using the same exact setting.

 

I did not find much comments about this articles on the internet, except for some criticism at a Whonix-Forum:

https://forums.whonix.org/t/strong-flow-correlation-attacks-on-tor-using-deep-learning/6010/4

 

What is your opinion? Does this article exaggerate things (like some previous studies obviously did)? And how relevant is it for VPN (opposed to TOR)?