Artificial Intelligence enabled investigation of biometric data using radar systems for advancements in assistive technologies

在这个项目中，我们的目标是开发一个人工智能超宽带(UWB)雷达系统，其目的是对人体生物特征数据进行非侵入性测量和预测。这类项目的主要应用是在养老院中用于检测严重和突发的健康问题。这项研究还探索了在封闭环境中检测多人的新方法。这样的系统测量和预测的数据无疑可以用于多种应用；例如，在工业环境中，自动感知特定区域内的人的存在是一个非常安全的问题。在商业环境中，可视摄像机数据是极大的数据安全问题来源，例如，了解多个人的存在和行为可以用来使周围环境更节能。该系统的主要用途是使用雷达技术测量和识别房间内一个或多个人的心率。使用这些信息和机器学习算法，该系统应该能够识别异常或任何其他重大问题，如心脏病发作或心脏完全停止跳动。它将被用作预防性警报系统，特别适用于特别容易出现突发医疗状况的个人，主要是老年人或婴儿。这个博士项目专门研究嵌入式系统，利用人工智能能力来分析天生微弱和噪声的雷达信号。本项目希望回答有关将机器学习用于信号处理的问题；主要是，是否有任何有效的方法来分析微弱的雷达信号来进行生物特征数据测量？雷达系统能被改进到可能探测到脑电波吗？为了试图解决这个问题，我们必须从整体上看待手头的问题，并认识到这个项目结合了计算机科学、电气工程和医学科学的领域。我们将首先开发雷达系统，这是获取特定数据不可或缺的一部分，然后改进硬件，使之能够进行测量。使用这些获得的信息，重要的是从医学角度理解数据，并将这种理解构建到可能开发用于分析的任何计算算法中。最后，必须对系统进行彻底的测试，以最大限度地提高其识别生物特征数据中的模式的效率。为了成功实施这一项目，我们必须开发新的信号处理算法，并将机器学习应用于这些算法。我们还必须确定从微弱和噪声的雷达信号中识别和处理生物特征数据的优化技术。最后，我们应该调查这个固定的雷达系统最终能够探测脑电波的可能性。这些是这个博士项目要解决的一些主要技术挑战。

In this project, we aim to develop an artificial intelligence enabled ultra-wideband (UWB) radar system, the purpose of which would be for the non-invasive measurement and prediction of human biometric data. The primary application for this type of project is for uses in elderly people's homes for the detection of severe and sudden health issues. This research also investigates into novel methods of detecting multiple people within closed environments. The data measured and predicted by such a system can undoubtedly be used for a multitude of applications; in industrial settings for example, automated awareness on the presence of people within specific areas is a matter of great security and safety. Within commercial environments, where visual camera data is a source of great data security concern, understanding the presence and behaviour of multiple people can, for example, be used to make the surroundings more energy efficient. The main and primary use of the system is to measure and recognise the human heart rate of an individual or individuals within a room using radar technology. Using this information, and machine learning algorithms, the system should then be able to recognise abnormalities or any other major issues, such as a heart attacks or the sudden stopping of the heart completely. This is to be used as a preventative alarm system for individuals particularly prone to sudden medical conditions, mainly speaking elderly people or infants.This PhD project is specifically looking into embedded systems utilising artificial intelligence capabilities to analyse inherently weak and noisy radar signals. This project hopes to answer questions related to using machine learning for the purpose of signal processing; mainly being, are there any effective methods of analysing weak radar signals for the biometric data measurements? And can a radar system be refined to possibly detect brain waves? To attempt to solve this problem, we must take a holistic view of the questions at hand and understand that this project combines areas of computer science, electrical engineering and the medical sciences. We will begin by developing the radar system, which is integral to the acquisition of the specified data and then refining the hardware to the point where measurements can be made. Using this acquired information, it is important to understand the data from a medical perspective and build this understanding into any computational algorithms that may be developed for analysis. Finally, thorough testing must be done on the system in order to maximise its efficiency in recognising patterns within the biometric data.To successfully carry out this project, we must develop novel signal processing algorithms and apply machine learning to them. We must also identify optimized techniques for identifying and processing biometric data from weak and noisy radar signals. Finally, we should investigate into the possibility of this stationary radar system being capable of ultimately detecting brain waves. These are some of the major technical challenges to be addressed by this PhD project.