| 58 | 0 | 11 |
| 下载次数 | 被引频次 | 阅读次数 |
针对基于WiFi信道状态信息(Channel State Information, CSI)的人体活动识别(Human Activity Recognition, HAR)技术在多人场景下，因动作遮挡与重叠导致识别困难的问题，探索提升复杂场景识别准确率与鲁棒性的方法。为此，提出时频多尺度金字塔网络(Time-Frequency Multi-Scale Pyramid Network, TFMS-Net)模型，通过频域增强模块实施多通道滤波，引入门控机制增强信号抗干扰能力；利用动态时空卷积模块结合双重门控机制捕捉动作特征，采用多尺度空洞卷积挖掘交互信息，借助通道分割策略并行处理子信号，通过分层下采样实现个体动作特征分离。在室内多人活动场景实验中，该模型对复杂动作的识别准确率超过92%,显著提升了基于WiFi CSI的HAR技术在多人场景下的鲁棒性，为非接触式多人动作识别提供了有效的解决方案。
Abstract：To address the challenge of Human Activity Recognition(HAR) based on WiFi Channel State Information(CSI) in multi-person environments, where motion occlusion and overlapping lead to recognition difficulties, methods to enhance recognition accuracy and robustness in complex scenarios are explored. A Time-Frequency Multi-Scale Pyramid Network(TFMS-Net) model is proposed, which employs a frequency-domain enhancement module to perform multi-channel filtering in the FFT domain and introduces a gating mechanism to enhance signal anti-interference capabilities. It utilizes a dynamic spatial-temporal convolution module combined with a dual gating mechanism to capture motion features, while adopting multi-scale dilated convolutions to mine interactive information. Additionally, it uses a channel segmentation strategy to process sub-signals in parallel and achieves individual motion feature separation through hierarchical down-sampling. Experiments in indoor multi-person activity scenarios show that the model achieves a recognition accuracy of over 92% for complex motions, significantly enhancing the robustness of WiFi CSI-based HAR technology in multi-person scenarios and providing an effective solution for non-contact multi-person motion recognition.
[1] SHAHBAZIAN R,TRUBITSYNA I.Human Sensing by Using Radio Frequency Signals:A Survey on Occupancy and Activity Detection[J].IEEE Access,2023,11:40878-40904.
[2] LI H J,HE X,CHEN X K,et al.Wi-Motion:A Robust Human Activity Recognition Using WiFi Signals[J].IEEE Access,2019,7:153287-153299.
[3] MA Y S,ZHOU G,WANG S Q.WiFi Sensing with Channel State Information:A Survey[J].ACM Computing Surveys,2019,52(3):1-36.
[4] KHAN M A,JAVED K,KHAN S A,et al.Human Action Recognition Using Fusion of Multiview and Deep Features:An Application to Video Surveillance[J].Multimedia Tools and Applications,2024,83(5):14885-14911.
[5] LI Y,YANG G C,SU Z D,et al.Human Activity Recognition Based on Multienvironment Sensor Data[J].Information Fusion,2023,91:47-63.
[6] ZHAO P J,LU C X,WANG B,et al.CubeLearn:End-to-End Learning for Human Motion Recognition from Raw Mmwave Radar Signals[J].IEEE Internet of Things Journal,2023,10(12):10236-10249.
[7] 司明豪，汪云甲，徐生磊，等.基于LOS/NLOS识别的WiFi FTM定位[J].无线电工程，2021,51(8):802-809.
[8] 李玉柏，孙迅.基于迁移学习提高 WiFi 室内定位中信道状态信息指纹库的鲁棒性[J].电子与信息学报，2023,45(10):3657-3666.
[9] 张梦，管同元，卓豪辉，等.基于胶囊网络的WiFi手势识别方法[J].无线电工程，2022,52(12):2288-2295.
[10] GUO Z X,XIAO F,SHENG B Y,et al.WiReader:Adaptive Air Handwriting Recognition Based on Commercial WiFi Signal[J].IEEE Internet of Things Journal,2020,7(10):10483-10494.
[11] ABDELNASSER H,YOUSSEF M,HARRAS K A.WiGest:A Ubiquitous WiFi-based Gesture Recognition System[C]//2015 IEEE Conference on Computer Communications.Hong Kong:IEEE,2015:1472-1480.
[12] LI T C,GAO S C,ZHU Y J,et al.SM-TCNNET:A High-performance Method for Detecting Human Activity Using WiFi Signals[J].Applied Sciences,2023,13(11):6443.
[13] NAKAMURA T,BOUAZIZI M,YAMAMOTO K,et al.Wi-Fi-CSI-based Fall Detection by Spectrogram Analysis with CNN[C]//2020 IEEE Global Communications Conference.Taipei:IEEE,2020:1-6.
[14] HU Y Q,ZHANG F,WU C S,et al.DeFall:Environment-independent Passive Fall Detection Using WiFi[J].IEEE Internet of Things Journal,2021,9(11):8515-8530.
[15] ALAZRAI R,AWAD A,ALI M Z,et al.An Approach for Recognizing Two-human Interactions Using Channel State Information[C]//2022 9th International Conference on Electrical and Electronics Engineering.Alanya:IEEE,2022:301-305.
[16] KABIR M H,RAHMAN M H,SHIN W.CSI-IANet:An Inception Attention Network for Human-Human Interaction Recognition Based on CSI Signal[J].IEEE Access,2021,9:166624-166638.
[17] XIAO N,YANG P L,YAN Y B,et al.Motion-Fi:Recognizing and Counting Repetitive Motions with Passive Wireless Backscattering[C]//IEEE Conference on Computer Communications.Honolulu:IEEE,2018:2024-2032.
[18] KHAN M M U,SHAMS A B,RAIHAN M S.A Prospective Approach for Human-to-Human Interaction Recognition from Wi-Fi Channel Data Using Attention Bidirectional Gated Recurrent Neural Network with GUI Application Implementation[J].Multimedia Tools and Applications,2024,83(22):62379-62422.
[19] KABIR M H,SHIN W.Deep Convolution Neural Network Based CSI Signal Classifier to Recognize Human-Human Interactions[C]//2021 IEEE International Conference on Consumer Electronics-Asia.Gangwon:IEEE,2021:1-3.
[20] GUO L L,ZHANG H,WANG C,et al.Towards CSI-based Diversity Activity Recognition via LSTM-CNN Encoder-Decoder Neural Network[J].Neurocomputing,2021,444:260-273.
[21] ALAZRAI R,AWAD A,BAHAA A,et al.A Dataset for Wi-Fi-based Human-to-Human Interaction Recognition[J].Data in Brief,2020,31:105668.
[22] LEI T,ZHANG D,WANG R S,et al.MFP-Net:Multi-scale Feature Pyramid Network for Crowd Counting[J].IET Image Processing,2021,15(14):3522-3533.
基本信息:
DOI：
中图分类号:TP391.41
引用信息:
[1]张亚军,张涛,李峰等.基于时频多尺度融合的CSI双人动作识别方法[J].无线电工程,2025,55(08):1598-1606.
基金信息:
新疆维吾尔自治区自然科学基金(2022D01C54); 新疆大学博士研究启动基金(202212120001)~~