基于移动通信网信号的人体动作识别研究

钱能; 陆城灵; 万雨轩; 吴哲夫

doi:10.11959/j.issn.1000-0801.2026015

您当前的位置：

首页 >

文章列表页 >

基于移动通信网信号的人体动作识别研究

更新时间：2026-01-21

- 基于移动通信网信号的人体动作识别研究
- Human action recognition based on mobile communication signals
- 电信科学 2026年页码：1-12
- 作者机构：
  
  1.浙江邮电职业技术学院电子与通信工程学院，浙江杭州 312000
  2.浙江工业大学网络空间安全研究院，浙江杭州 310000
- 作者简介：
  
  [ "钱能（1982- ）男，浙江邮电职业技术学院电子与通信工程学院副教授，主要研究方向为普适感知和信息融合技术等。" ]
  [ "陆城灵（1999- ）女，浙江工业大学网络空间安全研究院硕士生，主要研究方向为无线感知和深度学习技术等。" ]
  [ "万雨轩（1999- ）女，浙江工业大学网络空间安全研究院博士生，主要研究方向为通感一体化和信号处理技术等。" ]
  [ "吴哲夫（1971- ）男，浙江工业大学网络空间安全研究院副教授、博士生导师，主要研究方向为无线感知、信号处理和深度学习技术等。" ]
- 基金信息：
  
  浙江省自然科学基金重点项目(LZ22F010005)
- DOI：10.11959/j.issn.1000-0801.2026015
  中图分类号： TP393
- 收稿：2025-05-18，
  
  修回：2025-06-29，
  
  录用：2025-07-07，
  
  网络出版：2026-01-06，
- 稿件说明：
移动端阅览
钱能,陆城灵,万雨轩等.基于移动通信网信号的人体动作识别研究[J].电信科学,

QIAN Neng,LU Chengling,WAN Yuxuan,et al.Human action recognition based on mobile communication signals[J].Telecommunications Science,
钱能,陆城灵,万雨轩等.基于移动通信网信号的人体动作识别研究[J].电信科学, DOI：10.11959/j.issn.1000−0801.2026015.

QIAN Neng,LU Chengling,WAN Yuxuan,et al.Human action recognition based on mobile communication signals[J].Telecommunications Science, DOI：10.11959/j.issn.1000−0801.2026015.

摘要

Wi-Fi信道状态信息（channel state information，CSI）的人体活动识别方法（human activity recognition，HAR）在动作监测等领域被广泛应用，但仍面临高部署成本和感知范围有限等挑战。为解决这些问题，提出了一种移动通信网长期演进的注意力卷积神经网络（long term evolution attention-guided ConvNeXt ，LTE-ACN），利用移动信号进行活动识别。首先，基于小区参考信号提取信道状态信息有效特征；其次，进行噪声滤波、Savitzky-Golay平滑、峰谷增强和格拉姆角场变换等信号处理来构建动作数据集；最后，将增强后的图像数据输入LTE-ACN网络模型，通过改进的ConvNeXt架构有效地降低了特征信息损失，同时引入注意力机制模块（convolutional block attention module，CBAM）提升关键特征的表达能力，强化了空间域内的特征关联性。实验结果表明，所提方法的6类动作识别平均准确率达到了96.44%，验证了基于LTE信号进行人体动作识别的可行性。

Abstract

Wi-Fi channel state information (CSI)-driven human activity recognition (HAR) has been widely studied for applications in activity monitoring. However

challenges such as high deployment costs and limited sensing range remain. To address these

long term evolution attention-guided ConvNeXt (LTE-ACN) was proposed in this paper

which used mobile communication LTE signals for activity recognition. Firstly

effective features of channel state information were extracted based on cell reference signals. Then

signal processing methods including noise filtering

Savitzky-Golay smoothing

peak-valley enhancement

and Gramian angular field transformation were applied to construct the motion dataset. Finally

the enhanced image data was input into the LTE-ACN network model

in which the improved ConvNeXt architecture effectively reduced the feature information loss while the incorporated convolutional block attention module (CBAM) attention mechanism enhanced the expressive capability of key features and strengthened the feature correlations in the spatial domain. Experimental results demonstrate that the proposed method achieves an average accuracy of 96.44% in the six actions recognition

verifying the feasibility of LTE signal-based human motion recognition.

关键词

Keywords

references

NIKPOUR B , SINODINOS D , ARMANFARD N . Deep reinforcement learning in human activity recognition: a survey and outlook [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2025 , 36 ( 3 ): 4267 - 4278 .

CHEN Z , ZHU Q , SOH Y C , et al . Robust human activity recognition using smartphone sensors via CT-PCA and online SVM [J ] . IEEE Transactions on Industrial Informatics , 2017 , 13 ( 6 ): 3070 - 3080 .

ERTIN E , STOHS N , KUMAR S , et al . AutoSense: unobtrusively wearable sensor suite for inferring the onset, causality, and consequences of stress in the field [C ] // Proceedings of the 9th ACM Conference on Embedded Networked Sensor Systems . Seattle, USA : ACM , 2011 : 274 - 287 .

POLITI O , MPORAS I , MEGA V . Human motion detection in daily activity tasks using wearable sensors [C ] // Proceedings of the 22nd European Signal Processing Conference (EUSIPCO) . Lisbon : IEEE , 2014 : 2315 - 2319 .

AGGARWAL J K , RYOO M S . Human activity analysis: a review [J ] . ACM Computer Survey , 2011 , 43 ( 3 ): 1 - 43 .

ALRASHDI I , SIDDIQI M H , ALHWAITI Y et al ., Maximum entropy Markov model for human activity recognition using depth camera [J ] . IEEE Access , 2021 , 9 : 160635 - 160645 .

ABBAS M , JEANNÈS R L B . Exploiting local temporal characteristics via multinomial decomposition algorithm for real-time activity recognition [J ] . IEEE Transactions on Instrumentation and Measurement , 2021 , 70 ( 11 ): 1 - 11 .

XU W , LIU J , ZHANG S , et al . RFace: anti-spoofing facial authentication using COTS RFID [C ] // Proceedings of the IEEE INFOCOM 202l . Vancouver : IEEE , 2021 : 1 - 10 .

ZHENG T , CHEN Z , ZHANG S , et al . Catch your breath: simultaneous RF tracking and respiration monitoring with radar pairs [J ] . IEEE Transactions on Mobile Computing , 2023 , 22 ( 11 ): 6283 - 6296 .

ZHANG S , ZHENG T , CHEN Z , et al . Can we obtain fine-grained heartbeat waveform via contact-free RF-sensing? [C ] // Proceedings of the IEEE INFOCOM 2022 . London : IEEE , 2022 : 1759 - 1768 .

CHA J , CHOI I , YOO K . Human activity classification with user-held fmcw radar using deep recurrent neural networks [J ] . IEEE Sensors Letters , 2025 , 9 ( 5 ): 1 - 4 .

WU Z , XU Q , LI J , et al . Passive indoor localization based on csi and naive bayes classification [J ] . IEEE Transactions on Systems, Man, and Cybernetics: Systems , 2018 , 8 ( 9 ): 1566 - 1577 .

PAN Y , ZHOU Z , GONG W , et al . SAT: a selective adversarial training approach for wifi-based human activity recognition [J ] . IEEE Transactions on Mobile Computing , 2024 , 23 ( 12 ): 12706 - 12716 .

FENG Y , XIE Y , GANESAN D , et al . LTE-based pervasive sensing across indoor and outdoor [C ] // Proceedings of the SenSys 2021 . New York : ACM , 2021 : 138 - 151 .

SONNY A , RAI P K , KUMAR A , et al . Deep learning-based smart parking solution using channel state information in lte-based cellular networks [C ] // Proceedings of the International Conference on COMmunication Systems & NETworkS (COMSNETS) . Bengaluru , 2020 : 642 - 645 .

LIU Z , MAO H , WU C Y , et al . A ConvNet for the 2020 s[C ] // Proceedings of the CVF Conference on Computer Vision and Pattern Recognition (CVPR) . New Orleans : IEEE , 2022 : 11966 - 11976 .

MA Y , ZHOU G , WANG S , et al . SignFi: sign language recognition using WiFi [J ] . ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies , 2018 , 2 ( 1 ): 1 - 21 .

YOUSEFI S , NARUI H , DAYAL S , et al . A survey on behavior recognition using wifi channel state information [J ] . IEEE Communications Magazine , 2017 , 55 ( 10 ): 98 - 104 .

LEE H , AHN C R , CHOI N . Fine-grained occupant activity monitoring with Wi-Fi channel state information: practical implementation of multiple receiver settings [J ] . Advanced Engi neering Informatics , 2020 , 46 ( 10 ): 101147 .

SHI Z , ZHANG J A , XU R Y , et al . Environment-robust device-free human activity recognition with channel-state-information enhancement and one-shot learning [J ] . IEEE Transactions on Mobile Computing , 2022 , 21 ( 2 ): 540 - 554 .

ZHANG Y , CHEN Y , WANG Y , et al . CSI-based human activity recognition with graph few-shot learning [J ] . IEEE Internet of Things Journal , 2022 , 9 ( 6 ): 4139 - 4151 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于拓扑信息增强的异构图神经网络算法

基于多通道超图卷积与跨视图对比学习的会话推荐

基于GNN与注意力机制的文本分类模型

基于多尺度残差时间卷积网络的物联网入侵检测模型

面向协同感知的无人蜂群智能资源调度方案