SaTC: CORE: Small: Revolutionizing Wireless Inference Attacks and Defenses via Channel Analysis and Manipulation

Wireless signals are ubiquitous, invisible, and capable of penetrating obstacles. Due to the openness of the wireless medium, wireless communication is inherently vulnerable to eavesdropping. An eavesdropper may exploit intercepted wireless signals to infer sensitive information, such as keystrokes and private conversations. Unlike traditional keystroke or speech eavesdropping methods, which often require pre-installed malware on the victim’s device, wireless inference needs no such installation and does not depend on line-of-sight or physical proximity to the victim. However, existing wireless inference methods typically require a training process to map wireless measurements to specific typed or spoken content. These training-based approaches are often impractical, as it is difficult to collect generic training data about the victim and build a universal model. The project's novelties are addressing these limitations by transitioning wireless inference from training-required to training-free and developing proactive countermeasures to defend against wireless inference. The project's broader significance and importance are in creating a new research direction to combat wireless eavesdropping, enhancing the security of wireless systems, and advancing wireless security research. This project focuses on developing novel wireless eavesdropping techniques and corresponding counter-measures. This research aims to: (i) investigate new wireless keystroke and speech inference techniques via zero-training spatiotemporal analysis of wireless channel state measurements, and (ii) develop countermeasures against wireless inference, creating a new defense paradigm by leveraging wireless channel manipulation. This project creates techniques that improve the security of wireless channel-based recognition, identification, and authentication schemes. The results from this research will be openly shared. 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.