基于穿墙信道状态信息的行为识别方法

doi:10.19678/j.issn.1000-3428.0058816

摘要/Abstract

摘要： Wi-Fi CSI提供的被动式行为识别方法在许多场景得到了应用，但现有系统较少考虑信号穿墙的场景，导致信号穿墙后识别精度急剧下降。为提高系统的适应性，对信号穿墙后的行为识别方法进行研究，提出一种基于信道状态信息（CSI）的穿墙行为识别方法。根据CSI数据变化的特性，在预处理阶段，对CSI数据进行相位校准来证明原始CSI矩阵具有低秩性，并对其进行低秩矩阵分解，消除无用的静态CSI分量，凸显信号穿过墙壁后被掩盖的动态CSI分量。在识别行为阶段，利用时间反演算法解决CSI数据维度过高的问题，并简化计算。实验结果表明，与传统行为识别方法相比，该方法可大幅提升穿墙场景下的行为识别精度，在室内视距、室内非视距、穿墙的场景下平均识别精度分别可达94.1%、92.3%、90.7%。

关键词: 信道状态信息, 行为识别, 相位校准, 低秩矩阵分解, 静态分量消除, 时间反演算法

Abstract: The passive behavior recognition method provided by Wi-Fi CSI has been applied in many scenarios, but the existing systems hardly consider the scene where the signal passes through the wall, or the sharp decline in recognition accuracy after the signal passes through the wall. In order to improve the adaptability of the system, the method of recognizing the behavior of the signal passing through the wall is studied, and a recognition method based on Channel State Information(CSI) is proposed. According to the changing characteristics of the CSI data, in the preprocessing stage, the phase calibration of the CSI data is carried out, and then the low rank of the original CSI matrix is proved and decomposed by the low rank matrix to eliminate the useless static CSI components, so that the dynamic CSI components which are masked to a large extent after the signal passes through the wall are highlighted. In the recognition behavior stage, the time reversal algorithm is used to solve the problem that the dimension of CSI data is too high, and the calculation is simplified. Experimental results show that compared with traditional behavior recognition methods, the accuracy of behavior recognition in through-wall scene can be greatly improved. The average recognition accuracy can reach 94.1%, 92.3% and 90.7% respectively within the range of indoor visibility, beyond the range of indoor visibility and in through-wall scenes.

Key words: Channel State Information(CSI), behavior identification, phase calibration, decomposition of low-rank matrix, static component elimination, time inversion algorithm

中图分类号:

TP391

蒙倩霞, 余江, 常俊, 浦钰, 陈澄. 基于穿墙信道状态信息的行为识别方法[J]. 计算机工程, 2021, 47(8): 201-209.

MENG Qianxia, YU Jiang, CHANG Jun, PU Yu, CHEN Cheng. Behavior Recognition Method Based on Channel State Information in Through-the-Wall[J]. Computer Engineering, 2021, 47(8): 201-209.

https://www.ecice06.com/CN/Y2021/V47/I8/201

图/表 14

20210819200847

20210819200852

20210819200856

20210819200901

20210819200906

20210819200912

20210819200916

20210819200920

20210819200924

20210819200928

20210819200932

20210819200937

20210819200942

20210819200947

参考文献

[1] JAIAI A, KIM Y H, KIM Y J, et al. Robust human activity recognition from depth video using spatiotemporal multi-fused features[J]. Pattern Recognition, 2017, 61:295-308.
[2] BAYAT A, POMPLUN M, TRAN D A.A study on human activity recognition using accelerometer data from smartphones[J]. Procedia Computer Science, 2014, 34:450-457.
[3] JOSHI K, BHARADIA D, KOTARU M, et l.WiDeo:fine-grained device-free motion tracing using RF backscatter[C]//Proceedings of the 12th USENIX Symposium on Networked Systems Design and Implementation.Berkeley, USA:[s.n.], 2015:189-204.
[4] WANG Y, WU K, NI LM, et al. Wi-Fall:device-free fall detection by wireless networks[J]. IEEE Transactions on Mobile Computing, 2017, 16(2): 581-594.
[5] GU Y, LIU T, LI J.Eaosense:data-driven emotion sensing via off-the-shelf Wi devices[C]//Proceedings of 2018 IEEE International Conference on Communications.Washington D.C., USA:IEEE Press, 2018:842-855.
[6] DANG X C, ZHANG T, HAO Z J, et al. Wi-SD:a human motion recognition method based on CSI amplitude and phase information[C]//Proceedings of IEEE CWSN'19.Washington D.C., USA:IEEE Press, 2019:332-347.
[7] ZHENG X, WANG J, SHANG G L, et al. Smokey:ubiquitous smoking detection with commercial Wi-Fi infrastructures[C]//Proceedings of IEEE International Conference on Computer and Communications.Washington D.C., USA:IEEE Press, 2016:752-769.
[8] WANG W, LIU A X, SHAHZAD M, et al. Understanding and modeling of Wi-Fi signal based human activity recognition[C]//Proceedings of IEEE International Conference on Mobile Computing and Networking.Washington D.C., USA:IEEE Press, 2015:65-76.
[9] WANG W, LIU A X, SHAHZAD M.Gait recognition using Wi-Fi signals[C]//Proceedings of ACM International Joint Conference on Pervasive Ubiquitous Computer.New York, USA:ACM Press, 2016:363-373.
[10] LIU M, ZHANG L, YANG P, et al. Wi-Run:device-free step estimation system with commodity Wi-Fi[J]. Journal of Network and Computer Applications, 2019, 143(19): 77-88.
[11] CHOWDHURY T Z, LEUNG C, MIAO C, et al. Wi-HACS:leveraging Wi-Fi for human activity classification using OFDM subcarriers' correlation[C]//Proceedings of IEEE Global Conference on Signal and Information Processing.Washington D.C., USA:IEEE Press, 2017:338-342.
[12] YAN H, ZHANG Y, WANG Y, et al. Wi-Act:a passive WiFi-based human activity recognition system[J]. IEEE Sensors Journal, 2020, 20(1): 296-305.
[13] GAO Q, WANG J, MA X, et al. CSI-based device-free wireless localization and activity recognition using radio image features[J]. IEEE Transactions on Vehicular Technology, 2017, 66(11): 10346-10356.
[14] ZAIN U l, ABIDEN A, WANG H Y.Wi-based driver's activity monitoring with efficient computation of radio-image features[J]. Sensors, 2020, 20(5): 77-86.
[15] WANG Z, JIANG K, HOU Y, et al. A survey on CSI-based human behavior recognition in through-the-wall scenario[J]. IEEE Access, 2019, 7:772-793.
[16] 鲁翔.基于Wi-Fi信道状态信息的室内人员感知研究与实现[D]. 成都:电子科技大学, 2019. LU X.Research and implementation of indoor personnel perception based on Wi-Fi channel state information[D]. Chengdu:University of Electronic Science and Technology of China, 2019.(in Chinese)
[17] 张明军.空间复用MIMO系统的信号检测方法比较[J]. 长春师范大学学报(自然科学版), 2019, 38(3): 35-41. ZHANG M J.Comparison of signal detection methods for spatial multiplexing MIMO systems[J]. Journal of Changchun Normal University (Natural Science), 2019(6): 35-41.(in Chinese)
[18] 李晓薇, 余江, 常俊, 等. 一种基于CSI的非合作式人体行为识别方法[J]. 计算机科学, 2019, 46(12): 266-271. LI X W, YU J, CHANG J, et al. A non-cooperative human behavior recognition method based on CSI[J]. Computer Science, 2019, 46(12): 266-271.(in Chinese)
[19] WU X, CHU Z, YANG P.TW-See:human activity recognition through the wall with commodity Wi-Fi devices[J]. IEEE Transactions on Vehicular Technology, 2019, 68(1): 306-319.
[20] CANDES E J, LI X D, MA Y, et al. Robust principal component analysis[J]. Journal of the ACM, 2011, 58(3): 1-37.
[21] ZHU D, PANG N, LI G, et al. NotiFi:a ubiquitous Wi-Fi-based abnormal activity detection system[C]//Proceedings of 2017 International Joint Conference on Neural Networks.Washington D.C., USA:IEEE Press, 2017:1766-1773.

选择文件类型/文献管理软件名称

选择包含的内容