As cybersecurity breaches become an increasing threat to many in today’s technology-driven world, Nirnimesh Ghose, assistant professor at the School of Computing, is researching how fingerprinting can improve security — of devices, not people.
With a new grant from the National Science Foundation, Ghose will continue research in the expanding field of wireless fingerprinting, a physical-level authentication technique that plays a critical role in identifying individual technology devices.
Ghose will serve as principal investigator, collaborating with University of Cincinnati’s Boyang Wang on the project, which is jointly supported by the Secure and Trustworthy Cyberspace Program and the Established Program to Stimulate Competitive Research.
Radio fingerprinting can distinguish wireless devices using radio frequency signals because of the inherent device hardware imperfections inherent in such signals. Through their research, Ghose and his collaborators aim to develop new methods to promote the robustness, scalability and resilience of the wireless fingerprint through the synergy of deep learning and signal processing.
“In a wireless network connection, the verification server can verify a device’s wireless fingerprint. This prevents a malicious device from authenticating someone who compromised the password,” Ghose said. “It can also be used to detect malicious devices, such as B. Blacklisted UAVs attempting to forge credentials to evade detection.”
Ghose, whose research focuses primarily on network security, said that while wireless fingerprinting is more than a decade old, using machine learning for wireless fingerprinting is a recent innovation in the field.
“This has improved fingerprint performance, and we’re going to investigate how to make it more difficult for an adversary to forge a fingerprint,” Ghose said.
Ghose and his team will tackle the project by developing three new radio fingerprinting methods: designing complex-valued triplet neural networks; construction of physical layer supported generative adversarial networks; and building neural networks over the time domain, the frequency domain, and the time-frequency domain. Each new method will improve accuracy, detection, and resilience by targeting and protecting against different types of attacks.
According to Ghose, advances in wireless fingerprinting techniques could significantly improve the protection of the most vulnerable aspects of the technology.
“In general, endpoints are the weakest link in cyberspace, so this is an attempt to secure them,” Ghose said. “I look forward to finding new ways to make cyberspace safe.”
