Quote from: yAmAdA on July 28, 2014, 02:54:48 AM

Quote from: AnonyMint on July 27, 2014, 08:20:00 PM

Also I don't believe Monero is anonymous to the NSA and authorities with high reliability given the weaknesses in I2P and Tor.

I believe Tor and I2P should not be conflated. Timing attacks are more difficult against I2P. With Tor, you have exit nodes which make it easy to see one of the endpoints for the purpose of correlation. It can still be done without visible exits, but this is a bit harder.

If mining is centralized then the exit points are probably easy to find.

Quote from: yAmAdA on July 28, 2014, 02:54:48 AM

There has been some talk about introducing random delays to harden the mix network layer. Since that has various drawbacks, how about adding cover traffic instead? It is wasteful but should still allow low enough latencies while mitigating timing analysis. This is another thing that would not work well with an exit node based system.

I assume you mean relay nodes sending out dummy packets at random intervals, so latency doesn't increase for legitimate traffic (as long as relay nodes can handle the additional bandwidth).

I have not read the research on traffic analysis to comment with complete confidence. My technical understanding is likely close.

Seems to me the adversary could ignore all packets coming out of the exit nodes that didn't correlate with a low-latency to the targeted entry node. This is a statistical analysis. If it can be seen over time that a targeted entry node is always correlating with low-latency to one of the Monero exit nodes (made more easy to find by centralization of mining), then it is mathematically provable (within statistical confidence) that those packets are coming from the entry node. Thus IP obfuscation and anonymity broken.

The obvious solution of using a different entry node for each transaction sent, means you need to source many unregistered internet connections. Well then you might as well just use one unregistered internet connection then you don't need I2P/Tor and you can use Bitcoin or any coin.

And no improvements to onion routing (Chaum mix-nets) are reliable if the relay nodes are Sybil attacked, and many people assume they are because who is providing all this relay traffic for free. For example if 20% of the relay nodes are the adversary, then with 3 onion layer hops you have ≈1% (0.2^3=0.008) chance of being non-anonymous across the 3 hops each time you connect to the network. So over 100s of transactions your anonymity will be defeated. Worse yet, the research shows that the higher the % of the nodes the adversary can monitor (either by owning the node or by watching the routing of traffic across the node), then the frequency and/or randomization of latency of the cover traffic has to increase. Worse yet, even if you say your odds are low enough for your choice, when you lose anonymity then others in the ring signature lose anonymity too, i.e. your lower threshold choice cascades to others that wanted a higher threshold.

Thus you see low-latency Chaum mix-nets (onion routing) are a fundamentally flawed concept for anonymity against a global adversary. Even a hacker has access to a botnet might be able to Sybil attack the relay nodes.