IR-UWB Through-wall Radar Human Motion Status Dataset

IR-UWB Through-wall Radar Human Motion Status Dataset是由中南大学航空航天学院独立设计的IR-UWB穿墙雷达系统采集的数据集。该数据集包含4230条时间域范围剖面数据，每条数据像素为768×32，观测时间为4秒。数据集通过实际测量不同人体在不同穿透介质中的运动状态获得，主要用于研究人体运动状态的识别，如行走、站立静止等。数据集的创建旨在通过机器学习技术，特别是卷积神经网络（CNN），来评估和提高穿墙雷达在非接触人体信息检测和监控中的应用性能，解决现有算法评估缺乏公共数据集的问题。

The IR-UWB Through-wall Radar Human Motion Status Dataset was collected using an IR-UWB through-wall radar system independently designed by the School of Aeronautics and Astronautics, Central South University. This dataset contains 4230 time-domain range profile data samples, each with a size of 768×32 pixels and an observation duration of 4 seconds. It is obtained by actually measuring the motion states of different human subjects through various penetration media, and is primarily used for research on human motion state recognition, such as walking, standing still, etc. The creation of this dataset aims to evaluate and enhance the application performance of through-wall radars in non-contact human information detection and monitoring via machine learning technologies, particularly Convolutional Neural Networks (CNNs), and resolve the issue that existing algorithm evaluations lack public datasets.