Promising results in cyber security
For Chris Peikert, a specialist in theoretical computer science and a professor at the University of Michigan, the NSF-BSF partnership has provided many benefits beyond funding. He and his Israeli partner, Prof. Alon Rosen of the Herzliya Interdisciplinary Center, have received two grants as part of the NSF-BSF Cyber Security and Privacy Program.
“For me, it’s such an advantage just to have Alon as a partner. We met at MIT in 2005, when Alon was a postdoc and I was a graduate student. We started working together in lattice-based cryptography after discovering that we were both intrigued by some of the recent foundational works in the area. I might never have started on this topic that I love and have dedicated most of my career to without these initial collaborations!,” said Peikert, who is also a 2011 recipient of BSF’s Bergmann Memorial Award, given annually to promising young scientists applying for their first BSF grant.
Peikert credits the NSF-BSF grants with giving him opportunities to work with Rosen and his team.
“The NSF-BSF partnership has been a great way to formalize and support existing research relationships, and develop new ones, with these scientists,” Peikert said. “Israel is one of a few worldwide hubs of excellence in academic cryptography. It has excellent groups across several universities, and it also has a very strong technology and startup ecosystem for cryptography and security.”
Peikert focuses on lattice-based cryptography. Lattices are geometric objects that yield unique cryptographic and security properties. The field has growing interest in the computer world, since lattices appear to make cryptography more resistant to certain attacks, e.g., by quantum computers.
Rosen focuses on computational complexity theory. That is the practice of classifying computational problems according to their inherent difficulty. A problem is regarded as inherently difficult if the solution requires significant resources to fix.
Cyber attacks undoubtedly fall into this category. Today’s computers employ what’s known as symmetric-key encryption that enables them to encrypt information before it’s sent to another computer. It’s like a secret code that computers must know in order to decode information. This helps the transfer of information – but also makes computers vulnerable to attacks. This can impact everything from national security to identity theft. The consumer typing in a credit card number depends on symmetric key encryption as much as the Army field general communicating sensitive information to troops.
Based on what they have discovered so far, Peikert and Rosen believe that lattice-based cryptography can be an effective weapon against computer security breaches. They have built an efficient “collision-resistant hash function.” A hash function accelerates a computer user’s ability to look up information and verify its integrity. In cryptography, hash functions can be used to uniquely identify data. A collision-resistant hash function makes it very difficult for unauthorized users to substitute invalid data for correct data.
Peikert and Rosen’s efforts have created new private-key and public-key encryption systems that are proving to have solid security properties along with rich functionalities. Google has already conducted a successful test involving lattice-based cryptography, and interest is expected to grow as the challenge of computer security becomes more complex.
Last year, Peikert and Rosen received the Test of Time Award at the Theory of Cryptology Conference (TCC). They were honored for their work on a paper, Efficient Collision-resistant Hashing From Worst-case Assumptions on Cyclic Lattices, that resulted from their first collaboration in 2005. The paper was published by TCC in 2006, and the award recognizes papers presented at least eight years ago at the conference, that have made significant contributions to the cryptography field.
“We want our work to be vital to security and privacy in today’s networked world, and particularly for securing data on the cloud in a way that preserves the users’ utility from the encrypted data,” Rosen said. “We are very grateful to the NSF and BSF for helping us.”