Replicating Human Immune System via Harmonic Radar: A Framework and Preliminary Results in Thwarting Cyber-physical Attacks

Cyber-physical attacks target the intersection of digital systems and physical infrastructure. To mitigate such threats, we introduce a methodology based on harmonic radar and image classification, inspired by the human immune system’s self and non-self discrimination mechanism. We evaluate our framework with a use case focusing on authentication at the physical layer of USB devices—these latter devices being a common vector for cyber-physical attacks. Our approach achieves an average F1-score of 0.945 without requiring physical contact or activation to distinguish self from non-self devices. The results demonstrate the feasibility of the framework for reducing the attack surface in cyber-physical environments. This work lays the foundation for broader applications of harmonic radar in intrusion detection and hardware authentication.