Vulnerability and Possible Fix Found for Tor Network

[Guest_Jim_*]

Security and privacy are two points that have become very important to many Internet users in recent years, and are continuing to become more important all the time. One means to achieve these goals is to use the Tor network as a user and as a destination site. Researchers at MIT and the Qatar Computing Research Institute, however, have found ways to successfully attack the network, but also suggest means of addressing these issues.

The Tor network uses what is known as onion routing, where a request made by a user is hidden within layers of encryption. The request is then sent to a random computer in the Tor network, called a guard in this case, which removes one layer of encryption and passes on the request to another, random computer. This continues until the request is completely unencrypted and the final destination is connected to, and at this point no guards involved know both the destination and the sender. It is also possible to use the Tor network to hide sites, by creating introduction points that alone know the address of the destination. Once the user and host computers connect though, another Tor router is added to the circuit, as a private rendezvous point.

What the researchers discovered is that an attacker could, if it is acting as a guard along a chain, infer based on the packets it is routing, determine whether the circuit created is a web-browsing circuit, introduction-point circuit, or a rendezvous circuit. Using similar packet analysis, an attacker could also identify computers that are hosting a hidden site and determine what sites a user is accessing, all without breaking any encryption.

To protect against this the researchers recommend adding dummy packets to the sequences. These packets will serve to obfuscate the type of circuit being made, as each will look similar to the other. However more work still needs to be done to ensure this suggestion will indeed fix the issue.

Source: MIT

Back to original news post