CAREER: 4D mm-Wave Compressive Sensing and Imaging at One Thousand Volumetric Frames per Second

毫米波传感和成像系统在广泛的应用中无处不在，例如用于预测天气的地球大气探测，用于检测机场检查站潜在威胁的安全监控，以及用于伤口诊断和愈合的浅表组织的生物成像。当场景动态变化不迅速时，这些系统通常运行良好。不幸的是，在新兴的具有社会重要性的应用中，情况并非如此，比如在救援任务中成群结队的无人机，在道路上行驶的智能自动驾驶汽车，或者在移动中搜索自杀式炸弹袭击者的网络物理系统。该项目将通过开发第一个在快速变化的场景下运行的四维（4D）毫米波成像系统，使我们的社会受益，在这种情况下，必须快速做出安全关键决策。该系统的新应用之一将是在购物中心、体育场馆和办公楼等开放区域发现隐藏在衣服下或背包内的安全威胁。具体来说，该系统将能够扫描在26立方米的体积区域内移动的多人，每秒产生1000张三维图像帧，从而优于目前在机场检查站使用的现有毫米波传感和成像系统。除了社会影响之外，首席研究员（PI）将建立一个强大的教育计划，通过该计划，不同的观众可以了解基于波的成像系统的原理和局限性。通过与东北大学（NEU）合作教育计划合作，为学生开发新的课程和研究培训方法，以及通过制定学生向工业过渡的路线图，实现研究和教育的整合。外展计划包括与NEU的科学、技术、工程和数学（STEM）中心合作，为K-12、本科和代表性不足的学生提供研究经验，以及通过在线材料和公共场所进行教育。本CAREER项目的研究目标是了解使用4D（时空）编码的自适应压缩感知和成像系统的理论原理和基本局限性，并通过一种新型的4D毫米波自适应压缩成像雷达系统开发和实验验证这些原理。该系统将产生超过每秒1000帧的4D体帧率，每帧超过100万像素。实现4D自适应成像系统的主要挑战如下：1)系统必须能够处理可变动力学，即以不同速度移动的物体和位于不同焦距的物体；2)系统必须在有限的时间内采样具有足够信噪比的数据；3)系统必须以极快的速度采样数据，以实现高体积帧率的快速4D视频重建。该项目将解决以下挑战：(i)它将发展一种新的理论，将功能分析、信息论、压缩感知和适应性超材料结合在一起，以提高传感系统的信息传递效率；（ii）基于压缩成像系统期望的信息率和能量效率，开发新的数学框架来优化四维码；（iii）该系统将利用空间光调制器、涡流元透镜和压缩反射器来执行编码并动态适应成像区域的状态。本研究结果将通过提高成像性能、可靠性和效率，同时降低系统的硬件复杂性、总体成本和能耗，为提出的新型传感和成像系统奠定科学基础。

Millimeter-wave sensing and imaging systems are used ubiquitously for a wide range of applications, such as atmospheric sounding of the earth to forecast the weather, security monitoring to detect potential threats at airport checkpoints, and biological imaging of superficial tissues for wound diagnosis and healing. These systems typically operate well when the scene dynamics do not change rapidly. Unfortunately this is not the case in emerging societally-important applications like swarms of drones in rescue missions, smart self-driving cars on roadways, or cyber-physical systems searching for suicide bombers when they are on the move. This project will benefit our society with the development of the first four-dimensional (4D) millimeter-wave imaging system operating in fast changing scenarios, in which safety-critical decisions must be made quickly. One of the new applications of this system will be finding security threats, concealed under clothing or inside backpacks, in open areas like shopping malls, sport venues, and office buildings. Specifically, the system will have the capability to scan multiple people moving within a volumetric region of 26 cubic meters, producing 1000 image frames per second in three dimensions, thus outperforming existing millimeter-wave sensing and imaging systems that are currently used at airport checkpoints. In addition to the societal impact, the Principal Investigator (PI) will build a strong educational program through which diverse audiences can understand the principles and limitations of wave-based imaging systems. The integration of research and education will be accomplished through the development of new curricula and research training methods for students, as well as through the elaboration of a roadmap for transitioning students into industry, in collaboration with Northeastern University (NEU) Cooperative Education Program. The outreach plan includes enabling research experiences for K-12, undergraduate, and underrepresented students in collaboration with the Science, Technology, Engineering, and Mathematics (STEM) centers at NEU, as well as education through online materials and public venues.The research goal of this CAREER program is to understand the theoretical principles and fundamental limitations of adaptable compressive sensing and imaging systems using 4D (temporal and spatial) coding and to develop and experimentally validate these principles through a novel 4D millimeter-wave adaptive compressive imaging radar system. This system will produce 4D volumetric frame rates beyond 1000 frames per second, each frame having over one million pixels. The primary challenges of implementing 4D adaptable imaging systems are the following: 1) the system must be capable of handling variable dynamics, i.e., objects moving with different velocities and located at different focal ranges; 2) the system must sample data with sufficient signal to noise ratio during the limited period of time; and 3) the system must sample data extremely fast to perform fast 4D video reconstruction with high volumetric frame rates. This project will address these challenges as follows: (i) it will develop a new theory that brings together functional analysis, information theory, compressive sensing, and adaptable metamaterials to enhance the information transfer efficiency of sensing systems; (ii) it will develop a new mathematical framework to optimize 4D codes based on the desired information rate and energy efficiency of the compressive imaging system; and (iii) the system will utilize spatial light modulators, vortex-meta-lenses, and compressive reflectors to perform the coding and to dynamically adapt to the state of the imaging region. The result of this research will establish the scientific basis for the proposed new sensing and imaging systems, by enhancing the imaging performance, reliability, and efficiency while reducing the hardware complexity, overall cost, and energy consumption of the system.