Overlay anonymity systems like Tor are effective against many kinds of attacks on privacy, but they have significantly slower network performance than regular Internet traffic. The purpose of this project is to explore the design of anon.next, an anonymity system for the next-generation Internet. In anon.next, we embed anonymizing proxies into new Internet architectures, so that the network itself can provide efficient and effective privacy protection in a way that overlay designs cannot. This project seeks to make major advances in two areas key to the design of anon.next. First, methods to construct high-throughput paths of anonymizing proxies to route the user's packets to their destinations with minimal leakage of privacy. Second, secure methods to locate those proxies without relying on centralized directory servers or exposing the users' packet routes. Both components require us to develop new metrics for the privacy provided by a given anonymizing path and by the system as a whole. We are conducting analysis and simulation studies and validating these with extensive experimentation on GENI testbeds. These efforts will provide major insights into the design of anonymity systems, which provide personal privacy as well as censorship resistance and protection for whistle-blowers, journalists, and intelligence services. Our project is also part of the large effort required to understand and design distributed systems in the next-generation Internet. This project involves students at many levels, including middle and high school students in summer camp settings, undergraduate students in research into finding new attacks against our systems, and graduate students.
