Re: Burning bridges was Re: [i2p] I2P's security from global passive traffic analysis

Matthew Toseland <toad-EI5O+8PHWbJeeLb3ft/[email protected]>
Newsgroups gmane.network.freenet.general,gmane.network.i2p
Message-ID <[email protected]>
On Thu, Oct 27, 2005 at 01:37:28PM -0400, [email protected] wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> (continuing the reply w/ regards to the other non-CBR issues)
> 
> > This is as opposed to a system such as I2P operates now, where
> > connections are set up on the whim of the tunnel creator, where
> > passive traffic analysis can probably reveal where the tunnel is
> > going to, unless either its construction is so slow that it's
> > impossible to tell it from the other tunnels 
> 
> I wouldn't say construction, but operation.  Construction is just
> one packet, and payload does not immediately follow.

Hmmm. Well the idea I was trying to get across above was tied to "new
data flows", but basically, on each node, it must be plausible that a
packet going to a given node came from the local node or is forwarded
from any of several connected nodes which sent us data recently.
> 
> > (this is a serious option; users may not mind it if we give them
> > a nice GUI),

I was thinking of connection setup, or new traffic, here. Which can be
distinguished from existing flows if there is insufficient cover
traffic. Which I submit there always will be on an any-to-any network,
because most of the time a new connection flow will be A -> B -> C,
where A, B and C were not exchanging data before; it can be easily
distinguished from background traffic.
> 
> True, throttling individual tunnels to no more than X KBps has real
> potential for some applications.  I2P 3.0 will offer this [1]
> 
> [1] http://www.i2p.net/todo#batching
> 
> > or we are very lucky with cover traffic.
> 
> Tunnel creation itself doesn't require us to be very lucky, the
> attack P = (c/n)^h is negligible.  Tunnel operation against global
> passive adversaries, however, is a bitch.

It's a different c than in most of the other attacks though. If h is 3
and c is say all the nodes in the USA (via CALEA), then you have a
problem.
> 
> =jr
-- 
Matthew J Toseland - toad-EI5O+8PHWbJeeLb3ft/[email protected]
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.

_______________________________________________
chat mailing list
[email protected]
Archived: http://news.gmane.org/gmane.network.freenet.general
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/chat
Or mailto:[email protected]?subject=unsubscribe
signature.asc (application/pgp-signature, 189 B)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFDYRIuHzsuOmVUoi0RAithAJ99HHtbO9k8c39bQFPzjeN4WmsBAwCgpY17
eJcTaWywEpZ3siKu3ixe4wg=
=XYz/
-----END PGP SIGNATURE-----
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.