基于CSI实例标准化的域泛化人体动作识别模型

王杨; 许佳炜; 王傲; 夏慧娟; 赵传信; 季一木

doi:10.11959/j.issn.1000-436x.2024110

您当前的位置：

首页 >

文章列表页 >

基于CSI实例标准化的域泛化人体动作识别模型

学术论文 | 更新时间：2024-08-08

- 基于CSI实例标准化的域泛化人体动作识别模型
- Domain-generalization human activity recognition model based on CSI instance normalization
- 通信学报 2024年45卷第6期页码：196-209
- 作者机构：
  
  1.安徽师范大学计算机与信息学院，安徽芜湖 241002
  2.南京邮电大学计算机学院，江苏南京 210003
- 作者简介：
  
  [ "王杨（1971- ），男，安徽灵璧人，博士，安徽师范大学教授，主要研究方向为人工智能、物联网、增强现实。" ]
  [ "许佳炜（1996- ），男，安徽合肥人，安徽师范大学硕士生，主要研究方向为无线感知、表示学习。" ]
  [ "王傲（1999- ），男，安徽阜阳人，安徽师范大学硕士生，主要研究方向为机器学习。" ]
  [ "夏慧娟（1997- ），女，安徽庐江人，安徽师范大学研究员，主要研究方向为机器学习、信号处理、计算机视觉等。" ]
  [ "赵传信（1977- ），男，安徽凤阳人，博士，安徽师范大学教授，主要研究方向为无线可充电传感器网络、智能信息处理。" ]
  [ "季一木（1978- ），男，安徽无为人，南京邮电大学教授、博士生导师，主要研究方向为云计算、大数据、物联网和人工智能等。" ]
- 基金信息：
  
  国家自然科学基金资助项目(61871412);江苏省重点研发计划基金资助项目(BE2023004-2);安徽省自然科学基金重点项目(KJ2019A0938;KJ2021A1314;KJ2019A0979)
- DOI：10.11959/j.issn.1000-436x.2024110
  中图分类号： TP391
- 收稿：2024-03-27，
  
  修回：2024-05-28，
  
  纸质出版：2024-06-25
- 稿件说明：
移动端阅览
王杨,许佳炜,王傲等.基于CSI实例标准化的域泛化人体动作识别模型[J].通信学报,2024,45(06):196-209.

WANG Yang,XU Jiawei,WANG Ao,et al.Domain-generalization human activity recognition model based on CSI instance normalization[J].Journal on Communications,2024,45(06):196-209.
王杨,许佳炜,王傲等.基于CSI实例标准化的域泛化人体动作识别模型[J].通信学报,2024,45(06):196-209. DOI： 10.11959/j.issn.1000-436x.2024110.

WANG Yang,XU Jiawei,WANG Ao,et al.Domain-generalization human activity recognition model based on CSI instance normalization[J].Journal on Communications,2024,45(06):196-209. DOI： 10.11959/j.issn.1000-436x.2024110.

摘要

为了实现完全不依赖目标域数据的Wi-Fi跨域人体动作感知，提出了一种基于CSI实例标准化的域泛化人体动作识别模型INDG-Fi。INDG-Fi使用实例标准化去除CSI特征表示的领域信息，接着构建共享特征提取的动作分类器和域分类器，并通过动作偏向学习和对抗性的域学习，将编码层提取的特征偏向人体动作引起的信号特征，同时远离领域信号影响。为了让模型关注受人体动作影响更显著的子载波信号，在编码层中加入子载波注意力模块。实现结果表明，所提INDG-Fi在不可见的用户和位置的感知性能分别为97.99%和92.73%，能够实现鲁棒的跨域感知。

Abstract

To achieve Wi-Fi cross-domain human activity perception that was not dependent on target domain data

a domain-generalization human activity recognition model based on CSI instance normalization called INDG-Fi was proposed. The instance normalization standardization was utilized to remove domain information from the representation of CSI features by INDG-Fi. Then action classifiers and domain classifiers were constructed for shared feature extraction. By employing activity bias learning and adversarial domain learning

the model biased the features extracted from the encoding layer towards signal variations caused by human actions while moving away from domain signals. To enhance the model’s focus on subcarrier signals that were more significantly influenced by human actions

a subcarrier attention module was incorporated into the encoding layer. The implemented results demonstrate that the proposed INDG-Fi achieves perceptual accuracies of 97.99% and 92.73% for unseen users and locations

respectively

thus enabling robust cross-domain perception.

关键词

Keywords

references

CHEN Z H , ZHANG L , JIANG C Y , et al . WiFi CSI based passive human activity recognition using attention based BLSTM [J ] . IEEE Transactions on Mobile Computing , 2019 , 18 ( 11 ): 2714 - 2724 .

YAO Y C , ZHAO C X , PAN Y H , et al . Human gesture recognition based on CT—a hybrid deep learning model in Wi-Fi environment [J ] . IEEE Sensors Journal , 2023 , 23 ( 22 ): 28021 - 28034 .

TAN S , YANG J , CHEN Y Y . Enabling fine-grained finger gesture recognition on commodity WiFi devices [J ] . IEEE Transactions on Mobile Computing , 2022 , 21 ( 8 ): 2789 - 2802 .

LIU S J , CHEN Y R , WANG H , et al . A low-calculation contactless continuous authentication based on postural transition [J ] . IEEE Transactions on Information Forensics and Security , 2022 , 17 : 3077 - 3090 .

周牧 , 龙玥辛 , 蒲巧林 , 等 . 基于CSI张量分解的室内Wi-Fi指纹定位方法 [J ] . 通信学报 , 2021 , 42 ( 11 ): 159 - 171 .

ZHOU M , LONG Y X , PU Q L , et al . Indoor Wi-Fi fingerprint localization method based on CSI tensor decomposition [J ] . Journal on Communications , 2021 , 42 ( 11 ): 159 - 171 .

WU D , ZENG Y W , GAO R Y , et al . WiTraj: robust indoor motion tracking with WiFi signals [J ] . IEEE Transactions on Mobile Computing , 2023 , 22 ( 5 ): 3062 - 3078 .

YIN G L , ZHANG J Q , SHEN G X , et al . FewSense, towards a scalable and cross-domain Wi-Fi sensing system using few-shot learning [J ] . IEEE Transactions on Mobile Computing , 2024 , 23 ( 1 ): 453 - 468 .

ZHANG X , TANG C P , YIN K , et al . WiFi-based cross-domain gesture recognition via modified prototypical networks [J ] . IEEE Internet of Things Journal , 2022 , 9 ( 11 ): 8584 - 8596 .

ZHANG Y , ZHENG Y , QIAN K , et al . Widar3.0: zero-effort cross-domain gesture recognition with Wi-Fi [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2022 , 44 ( 11 ): 8671 - 8688 .

WANG J D , LAN C L , LIU C , et al . Generalizing to unseen domains: a survey on domain generalization [J ] . IEEE Transactions on Knowledge and Data Engineering , 2023 , 35 ( 8 ): 8052 - 8072 .

LIU S J , CHEN Z H , WU M , et al . WiSR: wireless domain generalization based on style randomization [J ] . IEEE Transactions on Mobile Computing , 2024 , 23 ( 5 ): 4520 - 4532 .

GAO R Y , ZHANG M , ZHANG J , et al . Towards position-independent sensing for gesture recognition with Wi-Fi [J ] . Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies , 2021 , 5 ( 2 ): 61 .

NIU K , ZHANG F S , WANG X Z , et al . Understanding WiFi signal frequency features for position-independent gesture sensing [J ] . IEEE Transactions on Mobile Computing , 2022 , 21 ( 11 ): 4156 - 4171 .

GAO R Y , LI W W , XIE Y X , et al . Towards robust gesture recognition by characterizing the sensing quality of WiFi signals [J ] . Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies , 2022 , 6 ( 1 ): 11 .

SHENG B Y , XIAO F , SHA L T , et al . Deep spatial-temporal model based cross-scene action recognition using commodity WiFi [J ] . IEEE Internet of Things Journal , 2020 , 7 ( 4 ): 3592 - 3601 .

BU Q R , YANG G , MING X X , et al . Deep transfer learning for gesture recognition with WiFi signals [J ] . Personal and Ubiquitous Computing , 2022 , 26 ( 3 ): 543 - 554 .

JIANG W J , MIAO C L , MA F L , et al . Towards environment independent device free human activity recognition [C ] // Proceedings of the 24th Annual International Conference on Mobile Computing and Networking . New York : ACM Press , 2018 : 289 - 304 .

KANG H , ZHANG Q , HUANG Q Y . Context-aware wireless-based cross-domain gesture recognition [J ] . IEEE Internet of Things Journal , 2021 , 8 ( 17 ): 13503 - 13515 .

ZOU H , YANG J F , ZHOU Y X , et al . Joint adversarial domain adaptation for resilient WiFi-enabled device-free gesture recognition [C ] // Proceedings of the 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA) . Piscataway : IEEE Press , 2018 : 202 - 207 .

SHI Z G , 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 .

ZHOU Z P , WANG F , YU J H , et al . Target-oriented semi-supervised domain adaptation for WiFi-based HAR [C ] // Proceedings of the IEEE INFOCOM 2022-IEEE Conference on Computer Communications . Piscataway : IEEE Press , 2022 : 420 - 429 .

HALPERIN D , HU W J , SHETH A , et al . Tool release: gathering 802.11n traces with channel state information [J ] . ACM SIGCOMM Computer Communication Review , 2011 , 41 ( 1 ): 53 .

CUI W , ZHANG L , LI B , et al . Received signal strength based indoor positioning using a random vector functional link network [J ] . IEEE Transactions on Industrial Informatics , 2018 , 14 ( 5 ): 1846 - 1855 .

ZHANG Y , QU C , WANG Y J . An indoor positioning method based on CSI by using features optimization mechanism with LSTM [J ] . IEEE Sensors Journal , 2020 , 20 ( 9 ): 4868 - 4878 .

浏览量

919

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于原型网络的跨环境通道状态信息人体活动识别模型改进

基于SFNet的大规模MIMO系统的CSI反馈算法

基于CSI张量分解的室内Wi-Fi指纹定位方法

快时变FDD大规模MIMO系统智能CSI反馈方法

基于信道状态信息的无源室内人员日常行为检测方法