“We were looking for any radio frequency emitter making noise within a confined space,” said Alyse Jones, a research associate in the institute’s Spectrum Dominance Division. “The key challenge was not just identifying if the signals were present, but determining which ones were relevant and needed to be looked into.”
Jones has been working on the SCISRS project for three years, starting when she was a graduate student at Virginia Tech.
“This was an amazing experience for me to get as a student,” Jones said. “As I became full time faculty, I was able to grow with the project and to transition into the areas that I have more interest and expertise in.”
In the second phase, the focus shifted to investigating changed or mimicked signals, such as Wi-Fi or Bluetooth, that look harmless but are actually carrying anomalous information. The team also delved into specific emitter identification, using machine learning to tag the physical hardware that sent a particular signal.
“Imagine having multiple identical radios transmitting the same message,” said William “Chris” Headley, associate director of the institute’s Spectrum Dominance Division. “Our goal was to differentiate between them, even if they all appear identical on the surface.”
Now entering the third phase, the project will add the detection of unintended electronic emissions — known as emanations — to its suite of algorithms. These emanations, such as those from common technology like monitors or keyboards, could unintentionally leak secure information.
“It’s like a game of cat and mouse,” said Michaels, who is also a professor of electrical and computer engineering. “The cat is our signal detector, hunting for the mouse. But with emanations, it’s more like finding a chewed-up bag of chips — the remnants tell you that something was there, even if you don’t see the mouse itself.”
