CRII: SaTC: Securing Smart Devices with AI-Powered mmWave Radar in New-Generation Wireless Networks

未来的网络系统，例如6 G蜂窝和未来的Wi-Fi网络，被设想为连接数十亿个异构智能设备，并使用毫米波（mmWave）技术实现高速和低延迟通信。该项目不是专注于使用毫米波进行数据通信，而是概述了一项具有挑战性的研究计划，旨在探索，实施和评估用于网络安全应用的新型毫米波传感技术。该项目的创新之处在于开发基于毫米波的生物识别技术，以抵御更高级的攻击，并更快地对潜在的数据泄露做出响应。该项目更广泛的意义和重要性在于，该研究将帮助我们对采用基于毫米波的传感技术进行潜在的安全应用有一个基本的了解。研究的成功结果将导致新的生物识别技术，可以提供更强大的设备和数据保护的访问控制。该项目包括三个部分。第一部分是通过分析用户身体的毫米波信号反射来开发mmLock，这是一种用户离开检测系统。它可以帮助自动锁定设备，以防止数据窃取时，用户是在不知不觉中远离他/她的设备。该研究将探索和解决同时具有多个移动对象的挑战性场景。第二部分通过检查从视觉分析中提取的用户头部运动模式与来自相应毫米波数据的模式之间的一致性来研究活力检测技术，以实现更安全的面部认证。它有可能抵御最先进的攻击，如3D投影攻击，并可以在任何照明条件下工作。第三项工作将实现一个原型系统，并评估其在安全性和可用性方面的性能。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Future networking systems, such as 6G cellular and future Wi-Fi networks, are envisioned to connect billions of heterogeneous smart devices and enable high-speed and low-latency communications using millimeter wave (mmWave) technologies. Instead of focusing on using mmWave for data communication, this project outlines a challenging research plan on exploring, implementing, and evaluating novel mmWave-based sensing techniques for cyber security applications. The project’s novelties are the development of mmWave-based biometrics to defend against more advanced attacks and to respond faster to potential data breaches. The project's broader significance and importance are that the research will help us to gain a fundamental understanding on adopting the mmWave-based sensing techniques for potential security applications. A successful outcome of the research will lead to new biometrics techniques that can provide stronger access control for device and data protection.The project consists of three parts. The first part is to develop mmLock, a user-leaving detection system by analyzing the mmWave signal reflections from the user body. It can help lock device automatically to prevent data theft when the user is unknowingly away from his/her device. The research will explore and address the challenging scenario of having multiple moving objects simultaneously. The second part investigates a liveliness detection technique by checking the consistency between the user’s head movement pattern extracted from vision analysis and the pattern from the corresponding mmWave data for more secure facial authentication. It has the potential to defend against the most advanced attacks such as 3D projection attack and can work in any lighting conditions. The third effort will implement a prototype system and evaluate its performance with regards to security and usability.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.