郑州大学学报（工学版）2025，Vol.46Issue(6)：49-57,9.DOI:10.13705/j.issn.1671-6833.2025.03.021
基于CSI主成分分割的人体动作识别方法
Human Activity Recognition Method Based on CSI Principal Component Segmentation
摘要
Abstract
The traditional method of human activity recognition based on channel state information(CSI)suffers from issues such as input data redundancy and limited feature extraction.To address this,a human activity recogni-tion approach based on CSI principal components and a dual-layer sliding window mechanism was proposed.First-ly,autlier removal and noise reduction were performed on the amplitude the use of a dual-layer sliding window mechanism based on principal component analysis enabled activity segmentation of preprocessed CSI data to elimi-nate irrelevant information and enhance model training efficiency.Subsequently,spatial and temporal analysis of the CSI data was conducted using convolutional neural network and bidirectional gated recurrent unit,with the inte-gration of a multi-head attention mechanism to focus on key information for achieving high-precision recognition of human activities.Experimental validation was performed using the WiAR and BAHAR public datasets,demonstra-ting that the proposed method could effectively recognize various human activities in diverse environments,while re-ducing the data volume by 5%.The accuracy achieved on the WiAR dataset was 96.53%,indicating superior per-formance compared to existing methods.关键词
信道状态信息/活动分割/卷积神经网络/双向门控循环单元/多头注意力机制Key words
channel state information/activity segmentation/convolutional neural network/bidirectional gated re-current unit/multi-head attention mechanism分类
信息技术与安全科学引用本文复制引用
饶壮,丁大钊,王依菁..基于CSI主成分分割的人体动作识别方法[J].郑州大学学报（工学版）,2025,46(6):49-57,9.基金项目
河南省科技攻关计划项目(232102210045) （232102210045）
嵩山实验室重大科研项目(ZZK202403002) （ZZK202403002）
