本论文探索了利用无源雷达开展人类活动识别(Human Activity Recognition, HAR)的新方法,重点研究非侵入式的Wi-Fi信道状态信息(Channel State Information, CSI)数据。传统HAR方法常采用摄像头、可穿戴设备等侵入式传感器,存在隐私风险。本研究利用CSI的非侵入特性,采用脉冲神经网络(Spiking Neural Networks, SNN)解析人体运动引发的信号变化。这类网络与DeepProbLog等符号推理框架结合,提升了HAR系统的适应性与可解释性。SNN功耗更低,适配隐私敏感场景。实验结果表明,基于SNN的神经符号模型准确率较高,是多领域HAR应用的潜力替代方案。
英文摘要:
The thesis explores novel methods for Human Activity Recognition (HAR) using passive radar with a focus on non-intrusive Wi-Fi Channel State Information (CSI) data. Traditional HAR approaches often use invasive sensors like cameras or wearables, raising privacy issues. This study leverages the non-intrusive nature of CSI, using Spiking Neural Networks (SNN) to interpret signal variations caused by human movements. These networks, integrated with symbolic reasoning frameworks such as DeepProbLog, enhance the adaptability and interpretability of HAR systems. SNNs offer reduced power consumption, ideal for privacy-sensitive applications. Experimental results demonstrate SNN-based neurosymbolic models achieve high accuracy making them a promising alternative for HAR across various domains.
Neuro-Symbolic Fusion of Wi-Fi Sensing Data for Passive Radar with Inter-Modal Knowledge Transfer
Politecnico di Milano(米兰理工大学);University of Brescia(布雷西亚大学);DEVCOM Army Research Lab(DEVCOM陆军研究实验室);University of California, Los Angeles(加州大学洛杉矶分校);Air Force Research Laboratory(空军研究实验室)
Exploring Spatial-Temporal Representation via Star Graph for mmWave Radar-based Human Activity Recognition
School of Computer Science and Informatics, University of Liverpool(计算机科学与信息学系,利物浦大学);School of Computer Science and Engineering, Southeast University(计算机科学与工程系,东南大学);City University of Hong Kong(香港城市大学);Knight Foundation School of Computing and Information Sciences, Florida International University(Knight基金会计算与信息科学系,佛罗里达国际大学)
Generalization Theory for Through-the-Wall Radar Human Activity Recognition
School of Information and Electronics, Beijing Institute of Technology(信息与电子学院,北京理工大学);Key Laboratory of Electronic and Information Technology in Satellite Navigation, Beijing Institute of Technology(卫星导航电子信息技术重点实验室,北京理工大学)
Through-the-Wall Radar Human Activity Micro-Doppler Signature Representation Method Based on Joint Boulic-Sinusoidal Pendulum Model
School of Information and Electronics, Beijing Institute of Technology(北京理工大学信息与电子学院);Jiaxing Research Center of Beijing Institute of Technology(北京理工大学嘉兴研究院);Key Laboratory of Electronic and Information Technology in Satellite Navigation, Beijing Institute of Technology(北京理工大学卫星导航电子信息技术重点实验室)
ALERT Open Dataset and Input-Size-Agnostic Vision Transformer for Driver Activity Recognition using IR-UWB
Department of ICE, Hanbat National University, Daejeon, Republic of Korea(信息电子工程系,翰 bam国立大学);Department of ECE and INMC, Seoul National University, Seoul, Republic of Korea(电子工程系和智能纳米材料中心,首尔国立大学);Department of ECE and INMC, Seoul National University(电子工程系和智能纳米材料中心,首尔国立大学)