[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [tor-talk] Padding effective against simple passive end-to-end correlation attacks?

On Sat, Jan 26, 2013 at 8:49 AM, adrelanos <adrelanos@xxxxxxxxxx> wrote:

> Not exactly. I can imagine, that telling "let your Tor connection open
> as often and long as possible and transfer some dummy/cover traffic as
> often as possible before you do actions you really want to hide" could
> improve anonymity.

Running a Bitcoin node set to use Tor as a proxy, and also publishing its
listening port as a hidden service, will generate a constant stream of
intermittent, fairly low-bandwidth connections, which slowly change over
time.  You don't even have to use Bitcoin for anything, just passively
participate in the P2P network traffic that gets relayed.

One machine I have, there are 8 outgoing connections and 14 incoming hidden
service connections, resulting in about 15-20 Kbps up and down, with 35
open Tor circuits.

Now whether this kind of chaff makes any difference to a correlation attack
I have no idea =)

tor-talk mailing list