Internet security expert Amir Houmansadr at the University of Massachusetts Amherst recently received a five-year, $581,458 CAREER grant from the National Science Foundation to combat Internet censorship by analyzing current censor-circumvention systems and designing a model that will lead to new anti-blocking tools.
Current anti-censorship techniques, whether used by people simply trying to get on social media or by dissidents and human rights workers, are mostly based on trial and error rather than on solid theoretical foundations, says Houmansadr. This "cat and mouse" or "whack-a-mole" approach means they can be defeated by most state-level censors.
"Censorship resistance tools exist already to help people circumvent those who block free access to the Internet," he says. "But they have different levels of success against different types of censors, depending on the user's location and how aggressive the censoring government is. My goal is to make it much, much easier to bypass censorship no matter where you are in the world."
Houmansadr adds, "No one should or could claim that they can offer the ultimate solution in five years, especially because the current Internet doesn't allow it. But I want to advance the state of the art and improve the availability of effective techniques to help the good guys."
As one who has personally experienced cyber censorship, he points out, "It's not the safest topic in computer science to work on, but it's one of the most impactful. It's more than just research to me; I have a personal sense of how important open access and free communication are on the Internet."
Houmansadr, an assistant professor in the College of Information and Computer Sciences at UMass Amherst, will use game theory, network science, and information theory, among other strategies, to analyze specific censorship resistance systems, categorize them and develop a theoretical model from which to mount new approaches. For example, he will identify parties in "the game," assess their objectives, measure their capabilities and use game theory to evaluate what costs and benefits are involved.
One reason the work is challenging, he notes, is that the Internet of today was not designed either to help or hinder censorship. Opponents of Internet censorship therefore have had to come up with ways to make the social, financial or political costs of censorship too hot for governments to handle.
A technique called decoy routing can help. It forces censors who want to block access to a particular site to also block benign, non-forbidden websites at a risk of high social cost. "In order to block CNN, for example, the decoy forces them to also block a popular game site. This kind of pressure can change a government's decision; there are instances where we have seen this work," Houmansadr says.
Another approach, VPN proxies, allows users to bypass censorship, but they are then detectable and if the censor is very aggressive, using them can be dangerous. Houmansadr estimates that one of the most popular systems for circumventing censorship, known as Tor, has close to 2 million users per day around the world who represent an army of proxies run by global volunteers.
One important new approach the cyber security expert plans to use is to take advantage of emerging communication devices and techniques such as content caching, content delivery networks, mobility and cloud computing to design new censorship circumvention tools. "These things are changing the way people use the Internet," he points out. "A second part of this research will look at these new paradigms of Internet communication and use them to devise new, sustainable ways to get around censorship."
NSF's Faculty Early Career Development (CAREER) program offers its highest award in support of junior faculty who exemplify the role of teacher-scholars through outstanding research, excellent education and the integration of education and research in the context of the mission of their organizations.