Wagner Samuel, Alkasimi Ahmad, Pham Anh-Vu
Department of Electrical and Computer Engineering, University of California, Davis, Davis, CA 95616, USA.
Sensors (Basel). 2022 Apr 1;22(7):2718. doi: 10.3390/s22072718.
We employ a 77-81 GHz frequency-modulated continuous-wave (FMCW) millimeter-wave radar to sense anomalous vibrations during vehicle transport at highway speeds for the first time. Secure metallic containers can be breached during transport by means of drilling into their sidewalls but detecting a drilling signature is difficult because the large vibrations of transport drown out the small vibrations of drilling. For the first time, we demonstrate that it is possible to use a non-contact millimeter-wave radar sensor to detect this micron-scale intrusive drilling while highway-speed vehicle movement shakes the container. With the millimeter-wave radar monitoring the microdoppler signature of the container's vibrating walls, we create a novel signal-processing pipeline consisting of range-angle tracking, time-frequency analysis, horizontal stripe image convolution, and principal component analysis to create a robust and powerful detection statistic to alarm if drilling is present. To support this pipeline, we develop a statistical model combining the vibrating container and the random vibrations induced by vehicle movement to explore the robustness of the sensor's detection capabilities. The presented results strongly support the inclusion of a millimeter-wave radar vibration sensor into a transport security system.
我们首次采用77 - 81千兆赫调频连续波（FMCW）毫米波雷达，在车辆以高速公路速度行驶时感应异常振动。在运输过程中，安全金属容器的侧壁可能会被钻孔破坏，但检测钻孔信号很困难，因为运输过程中的大幅振动会掩盖钻孔的微小振动。我们首次证明，在高速公路行驶的车辆摇晃容器时，使用非接触式毫米波雷达传感器检测这种微米级的侵入性钻孔是可行的。通过毫米波雷达监测容器振动壁的微多普勒信号，我们创建了一个新颖的信号处理流程，包括距离 - 角度跟踪、时频分析、水平条纹图像卷积和主成分分析，以创建一个强大而稳健的检测统计量，用于在存在钻孔时发出警报。为支持该流程，我们开发了一个统计模型，将振动容器和车辆运动引起的随机振动结合起来，以探索传感器检测能力的稳健性。所呈现的结果有力地支持了将毫米波雷达振动传感器纳入运输安全系统。
