Location Adaptive Motion Recognition Based on Wi-Fi Feature Enhancement

被引:0
作者
Shi, Wei [1 ]
Duan, Meichen [1 ]
He, Hui [1 ]
Lin, Liangliang [1 ]
Yang, Chen [1 ]
Li, Chenhao [1 ,2 ]
Zhao, Jizhong [1 ]
机构
[1] Xi An Jiao Tong Univ, Dept Comp Sci & Technol, Xian 710049, Peoples R China
[2] Ant Rongxin Chengdu Network Technol Co Ltd, Chengdu 610040, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2023年 / 13卷 / 03期
关键词
human motion recognition; channel state information; multi-signal classification algorithm; wireless perception;
D O I
10.3390/app13031320
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Action recognition is essential in security monitoring, home care, and behavior analysis. Traditional solutions usually leverage particular devices, such as smart watches, infrared/visible cameras, etc. These methods may narrow the application areas due to the risk of privacy leakage, high equipment cost, and over/under-exposure. Using wireless signals for motion recognition can effectively avoid the above problems. However, the motion recognition technology based on Wi-Fi signals currently has some defects, such as low resolution caused by narrow signal bandwidth, poor environmental adaptability caused by the multi-path effect, etc., which make it hard for commercial applications. To solve the above problems, we first propose and implement a position adaptive motion recognition method based on Wi-Fi feature enhancement, which is composed of an enhanced Wi-Fi features module and an enhanced convolution Transformer network. Meanwhile, we improve the generalization ability in the signal processing stage to avoid building an extremely complex model and reduce the demand for system hardware. To verify the generalization of the method, we implement real-world experiments using 9300 network cards and the PicoScenes software platform for data acquisition and processing. By contrast with the baseline method using original channel state information(CSI) data, the average accuracy of our algorithm is improved by 14% in different positions and over 16% in different orientations. Meanwhile, our method has best performance with an accuracy of 90.33% compared with the existing models on public datasets WiAR and WiDAR.
引用
收藏
页数:17
相关论文
共 44 条
[1]  
Abdelnasser H, 2015, IEEE INFOCOM SER
[2]  
Adewopo V, 2023, Arxiv, DOI arXiv:2210.12527
[3]   New Trends and Advancement in Next Generation Mobile Wireless Communication (6G): A Survey [J].
Ali, Sher ;
Sohail, Muhammad ;
Shah, Syed Bilal Hussain ;
Koundal, Deepika ;
Abul Hassan, Muhammad ;
Abdollahi, Asrin ;
Khan, Inam Ullah .
WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2021, 2021
[4]   Human Activity Recognition Based on CSI fragment with Action-value Method [J].
Chen, Hongxin ;
Zhang, Yong ;
Yin, Yuqing ;
He, Fei .
2022 ASIA CONFERENCE ON ALGORITHMS, COMPUTING AND MACHINE LEARNING (CACML 2022), 2022, :448-455
[5]   APFNet: Amplitude-Phase Fusion Network for CSI-Based Action Recognition [J].
Duan, Pengsong ;
Li, Hao ;
Zhang, Bo ;
Cao, Yangjie ;
Wang, Endong .
MOBILE NETWORKS & APPLICATIONS, 2021, 26 (05) :2024-2034
[6]   Multi-Variations Activity Based Gaits Recognition Using Commodity WiFi [J].
Fei, Huan ;
Xiao, Fu ;
Han, Jinsong ;
Huang, Haiping ;
Sun, Lijuan .
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2020, 69 (02) :2263-2273
[7]  
Glorot X., 2010, 13 INT C ARTIFICIAL, P249
[8]   Wiar: A Public Dataset for Wifi-Based Activity Recognition [J].
Guo, Linlin ;
Wang, Lei ;
Lin, Chuang ;
Liu, Jialin ;
Lu, Bingxian ;
Fang, Jian ;
Liu, Zhonghao ;
Shan, Zeyang ;
Yang, Jingwen ;
Guo, Silu .
IEEE ACCESS, 2019, 7 :154935-154945
[9]   GenePrint: Generic and Accurate Physical-Layer Identification for UHF RFID Tags [J].
Han, Jinsong ;
Qian, Chen ;
Yang, Panlong ;
Ma, Dan ;
Jiang, Zhiping ;
Xi, Wei ;
Zhao, Jizhong .
IEEE-ACM TRANSACTIONS ON NETWORKING, 2016, 24 (02) :846-858
[10]  
Inayat U., 2021, P 2021 INT C INNOVAT, P1