Smartphone Imposter Detector System Provides Clever Defense Against Smartphone Theft
In the ever-evolving landscape of technology, the SRC funded project "Design and Security Verification of Next-Generation Open-Source Processors" stands as a testament to innovation and collaboration. This groundbreaking project, led by Princeton’s highly accomplished Professor Ruby B. Lee and dedicated SRC Research Scholar, Dr. Guangyuan Hu, has made significant strides in the realm of processor design and security.
The project, co-funded by the National Science Foundation (NSF) and SRC’s Hardware Security (HWS) research program, aimed to address the critical need for secure and efficient open-source processors. The team, which included then-graduate student Dr. Zecheng He, focused on developing advanced methodologies for the design and verification of these processors, ensuring they met the highest standards of security and performance.
Their pioneering work culminated in the issuance of US patent 12111898, titled "Devices and methods for smartphone impostor detection using behavioral and environmental data," in October 2024. This patent represents a significant advancement in smartphone security, utilizing innovative techniques to detect impostors based on behavioral and environmental cues. Impostors are attackers who take over a smartphone and gain access to the legitimate user’s confidential and private information. The team proposed a defense-in-depth mechanism to detect impostors quickly with Deep Learning algorithms enhanced with statistical distributions, achieving better detection accuracy than previous Machine Learning algorithms that required computation of multiple features. Unlike prior work, they considered protecting the privacy of a user’s behavioral (sensor) data by not exposing it outside the smartphone. They also designed and implemented the Smartphone Impostor Detector (SID) architecture to support real-time and self-contained impostor detection at a very low hardware cost and energy consumption compared to other recurrent neural network (RNN) accelerators.
Collaborations with liaisons from SRC member companies Arm, AMD, and Intel brought valuable industry insights and expertise. This enhanced the project's outcomes and ensured the developed technologies were aligned with industry needs.
The success of this project not only highlights the importance of interdisciplinary collaboration but also underscores the vital role of funding from organizations like NSF and SRC in driving technological advancements. Dr. Lee and Dr. Hu's contributions, along with the support from Arm, AMD, and Intel, have paved the way for more secure and reliable open-source processors.
View Professor Ruby B. Lee’s GRC Hardware Security project, “Design and Security Verification of Next-Generation Open-Source Processors” (2844.002), at https://app.pillar.science/projects/5234/overview.
Read the University press article here: https://innovation.princeton.edu/news/2021/ruby-lee-and-guangyuan-hu-tiny-ai-module-detecting-smartphone-theft-and-anomalous
This technology is available for licensing. Contact Princeton's Office of Technology Licensing at https://patents.princeton.edu/about-us/contact-us.
