Don’t Trust, Verify: Prof. Nathan Keller Strategizes a More Secure Cyber-Future
When Prof. Nathan Keller entered the Technion at the age of 16, his teachers immediately recognized that this was a young man who was going places. Eventually, both Keller and one of his mentors independently decided that Bar-Ilan University was the place to go. In 2012, Keller was recruited to BIU as a member of the mathematics department. Less than a year later, his former advisor at the Technion – Rabbi Prof. Daniel Hershkowitz – was named BIU’s tenth President.
“Prof. Hershkowitz was the dean of the Math Faculty when I was doing my Masters,” says Keller, 32, a Russian-born expert in cryptology and combinatorics who later went on to earn his PhD from Hebrew University. “Luckily, he still found the time to help me.”
Judging from the buzz created by Keller’s early academic work, luck had very little to do with it.
“In 2003, my colleagues and I discovered that second-generation cellphones were insecure – simple, relatively cheap equipment made it possible to listen in on any conversation, or even hijack phones altogether,” Keller says, adding that, after these findings were published, the story received coverage in media outlets including CNN and the New York Times. “Industry leaders accepted our analysis. However, because the fix required an antenna re-design costing tens of millions of dollars, the problem wasn’t addressed until the next generation of phones was unveiled, this time including protection based on our research.”
Keller focuses on examining the efficacy of codes designed to protect sensitive information. Sometimes, however, the “chinks in the armor” that Keller discovers can be used to shift protected information out of the shadows, and into the right hands.
“Five years ago, we discovered that car immobilizers can be hacked so that an attacker can acquire the password,” Keller says. “Practically speaking, not many car thieves would choose code breaking over breaking a window, but our results were of interest to law enforcement authorities. At the request of a European police force, we created an algorithm that can be used to open immobilized cars without being detected – a technique that can be helpful in intelligence gathering missions.”
According to Keller, cryptographic analysis is increasingly a matter of national security. “Until the mid-20th century, most cryptology research was devoted to protecting military secrets,” he says. “But in the age of cyber-terror, civilian infrastructure presents a very attractive military target. A computer virus that knocks out a country’s electrical grid can be more threatening than a tank division.”
Keller also works in combinatorics – a branch of mathematics related to the study of finite or countable discrete structures, which has implications for a wide range of applications.
“I study problems where many ‘inputs’ going into a complex system have to result in a single ‘output’, such as a yes or no answer,” Keller explains. “Relevant to everything from elections, to economics, to communication networks, our techniques help identify the path most likely to overcome ‘noise’, and to provide useful results.”
In his own career path, Keller has consistently balanced the professional and the personal. “I served in the IDF in the hesder yeshiva track, and, eventually, completed rabbinical ordination,” he says. “Married by the time I began my post-doc, I chose not to go abroad for religious reasons. I still collaborate with one of my post-doctoral hosts at the Weizmann Institute: Prof. Adi Shamir – winner of the Turing Award, considered the ‘Nobel’ of computer science.”
Having presented talks at Bar-Ilan University, Prof. Keller was familiar with the math department when he made his move. “When I was invited to join the BIU faculty, I didn’t hesitate – I knew it was a good match.”