glymr_darkmoon at ml1.net
Sun Apr 16 01:17:12 UTC 2006
-----BEGIN PGP SIGNED MESSAGE-----
i think it could probably help counter traffic analysis quite a bit if
say one out of ten packets is sent out of order, and would only
marginally increase latency and would not affect bandwidth at all.
some study would need to be done to determine exactly what proportion
of disordering of the scheduler is required to make timing attacks
less effective, or preferably if possible, near to impossible to perform.
> Perhaps the server could insert short (randomized) delays between
> recieving a request and relaying it.
> On 4/15/06, *Paul Syverson* <syverson at itd.nrl.navy.mil
> <mailto:syverson at itd.nrl.navy.mil>> wrote:
> On Sat, Apr 15, 2006 at 07:30:48PM +0200, Nils Vogels wrote:
> > On 4/15/06, 2600denver < 2600denver at gmail.com
> <mailto:2600denver at gmail.com>> wrote:
> > >
> > > Does this seem like a plausable solution to stop timing and
> > > attacks?
> > Yes it most probably does,
> Actually, no it doesn't
> And such attacks can be real
> > but it also effectively slows down the
> > entire Tor network bij 66%, at least when you are counting the
> > traffic in the bandwidth limits that you set for tor traffic.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
More information about the tor-talk