RII Track-4:NSF: Fundamentals of Creating Trustworthy Medical Cyber-Physical Systems Under EMI Attacks

点击翻译按钮获取中文摘要

The application of medical cyber-physical systems such as wireless insulin pumps, infant incubators, and pacemakers continues to rise. Researchers repeatedly showed how adversaries could deceive the perception of systems by manipulating sensor signals of medical cyber-physical systems prior to digitization with intentional electromagnetic (EMI) signals. This proposal aims to address the threats with systematic and general approaches by enhancing the hardware and software co-design. The outcomes of this project will enhance the understanding of the security and reliability of medical cyber-physical systems under intentional EMI attacks. The training and hands-on experience on cutting-edge technologies and the collaborations established via this program will allow the PI to establish a highly productive and reputable cyber-physical system security laboratory and contribute to the inter-disciplinary research at UL Lafayette. This EPSCoR grant will be used to mentor and train a female graduate student in cutting-edge model-building techniques that can be used in pure research and/or numerous industrial applications, thus, contributing to the imperative effort of increasing diversity in STEM fields. Lastly, the PI will incorporate the theories and experiments into her undergraduate/graduate course on cyber-physical system security.This Research Infrastructure Improvement Track-4 EPSCoR Research Fellows (RII Track-4) project would provide a fellowship to an Assistant professor and training for a graduate student at the University of Louisiana at Lafayette (ULL). The proposed studies will advance the methodology and design principles to ensure the real-time reliability of medical cyber-physical systems in the presence of intentional Electromagnetic Inference (EMI) attacks on analog medical sensors such as glucose sensors. This project will lay a solid foundation for reliable medical cyber-physical systems that rely on real-time sensing and control. The proposed approaches will advance understanding of designing secure and reliable systems with imperfect medical analog sensors. In this project, one task is to create new hardware units to limit the injected errors within a tolerable range by designing and implementing matched dummy circuits to identify EMI signals. In addition, the proposed studies will improve the understanding of how to utilize redundant sensor circuits in defenses against intentional EMI injection attacks. Another task is to create a physics-informed neural network-based framework to diagnose corrupted sensor data induced by intentional EMI attacks by solving equations and predicting the sensor value. It will compute a baseline (i.e., blood glucose or body temperature) to correct the corrupted sensor data.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.