Ji, Bo, 1982-Zhang, Jin2023-09-032023-09-032023-08http://hdl.handle.net/20.500.12613/8953In recent years, mmWave radar technology has gained significant prominence and found diverse applications in everyday life, including autonomous driving, health monitoring, and security checks. However, the increasing adoption of mmWave radars also brings forth notable security concerns that warrant attention and investigation. One prominent category of security threats revolves around eavesdropping. Malicious actors can exploit mmWave radar technology to intercept private data from legitimate users. For example, these radars can capture acoustic vibrations, enabling attackers to eavesdrop on private calls or online meetings. Moreover, attackers can extend their eavesdropping capabilities to encompass screens and air-gap devices, posing a severe threat to user privacy. Surprisingly, prior studies have not adequately addressed the detection and localization of such "spy" radars. The second security threat involves spoofing attacks, where attackers inject false target information into victim radars through the transmission of spoofing signals or the deployment of passive tags in specific locations. Existing spoofing attack systems often assume that attackers possess prior knowledge of the victim radar's mode and parameters, which can be obtained from publicly available resources such as user manuals. However, this assumption proves inadequate when dealing with commercial off-the-shelf (COTS) radars, as users have the ability to modify the radar's mode and parameters. Interference constitutes the third security threat. Interference can be considered an "unintended spoofing signal" that leads the radar to mistake the interfering signal as its own received signal, thereby generating ghost objects. Mitigating interference presents a significant challenge, with previous research focusing on random adjustments of radar parameters or coordination through a central controller. However, these approaches suffer from high costs, time-consuming procedures, and limited efficiency. Therefore, our primary objective is to develop an affordable radar coordination system. Additionally, it is observed that the complexity associated with implementing hardware-based parameter adjustments. To overcome this limitation, this dissertation proposes Radar-Immune, a system that empowers radars to autonomously cancel interference. With Radar-Immune, the radar does not require adjusting its parameters to avoid interference anymore. This dissertation presents several novel systems to address these security threats. Firstly, we design a spy radar detection and localization system that safeguards user privacy against eavesdropping attacks. Secondly, we propose a radar mode classification and parameter estimation system to enable practical spoofing attacks. Furthermore, we introduce a radar coordination system that significantly reduces the cost and time required for interference mitigation compared to existing approaches. Finally, we present an interference cancellation system that enables radars to autonomously cancel interference without requiring parameter adjustments. By developing these systems, this dissertation contributes to enhancing the security, privacy, and robustness of mmWave radar technology.158 pagesengIN COPYRIGHT- This Rights Statement can be used for an Item that is in copyright. Using this statement implies that the organization making this Item available has determined that the Item is in copyright and either is the rights-holder, has obtained permission from the rights-holder(s) to make their Work(s) available, or makes the Item available under an exception or limitation to copyright (including Fair Use) that entitles it to make the Item available.http://rightsstatements.org/vocab/InC/1.0/Computer scienceText154392023-08-24Qiu_temple_0225E_15439.pdf