CAREER: Combating Censorship from within the Network

University Of Colorado At Boulder, Boulder CO

Investigators

Abstract

Over half of the world’s Internet population live in countries that censor political, social, or religious content online, and authoritarian governments continue to design and deploy increasingly sophisticated censorship systems. Internet censorship is thus a significant and growing threat, making research into circumvention technologies an important priority. While in the past proxies and (Virtual Private Networks) VPNs could be used to circumvent censorship, today, censors use complex techniques to find and block existing circumvention resources. This project proposes new ways to study and combat Internet censorship around the world by adopting a perspective similar to the censor’s themselves, and leveraging Internet infrastructure in non-censoring countries. First, this infrastructure will be used to measure “normal” Internet traffic, so that circumvention proxies can learn what to mimic in order to camouflage their activity and evade detection. Second, this project will use those findings and the network infrastructure itself to create new kinds of proxies that are harder for censors to block. Studying and combatting censorship from the network perspective will help level the playing field between censors and circumvention tool developers, ultimately supporting Internet users’ autonomy and access to information globally, including the billions of people worldwide living in repressively censored regions. This work will use network taps deployed at (Internet Service Providers) ISPs and Internet Exchange Points (IXPs) in non-censoring countries to measure popular Internet protocols and how they are used in practice, develop novel ways to detect new and emerging forms of censorship, and explore how existing proxies and circumvention strategies can be improved using ISP data. Building on the growing Refraction Networking deployment, this work also investigates new ways that the ISP perspective can be used to combat censorship directly, such as improvements in Refraction transports to better resist active probing attacks from censors, implementing peer-to-peer proxy protocols that are harder for censors to detect and block, and finding ways to directly make censorship more expensive for governments in practice. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.