计算机科学 ›› 2020, Vol. 47 ›› Issue (11): 25-31.doi: 10.11896/jsjkx.200200044

• 智能移动身份认证 • 上一篇    下一篇

基于WiFi信号的轻量级步态识别模型LWID

周志一1, 宋冰2, 段鹏松1, 曹仰杰1   

  1. 1 郑州大学软件学院 郑州 450000
    2 河南警察学院网络安全系 郑州 450046
  • 收稿日期:2020-02-08 修回日期:2020-07-03 出版日期:2020-11-15 发布日期:2020-11-05
  • 通讯作者: 曹仰杰(caoyj@zzu.edu.cn)
  • 作者简介:zhou_zhi_yi@163.com
  • 基金资助:
    国家自然科学基金(61972092)

LWID:Lightweight Gait Recognition Model Based on WiFi Signals

ZHOU Zhi-yi1, SHONG Bing2, DUAN Peng-song1, CAO Yang-jie1   

  1. 1 School of Software Engineering,Zhengzhou University,Zhengzhou 450000,China
    2 Henan Police College network Security Department,Zhengzhou 450046,China
  • Received:2020-02-08 Revised:2020-07-03 Online:2020-11-15 Published:2020-11-05
  • About author:ZHOU Zhi-yi,born in 1993,postgra-duate.His main research interests include application of deep learning in the field of WiFi.
    CAO Yang-jie,born in 1976,Ph.D.His current research interests include computer vision and intelligent computing,artificial intelligence and high-perfor-mance computing.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China(61972092).

摘要: 身份识别作为普适计算和人机交互领域的重要研究内容,受到研究者的广泛关注。基于WiFi信号的传统身份识别方法虽然取得了较大的进展,但仍然面临分类能力弱、模型存储代价高、训练时间长等问题。对此,提出了基于多层神经网络的轻量级步态识别模型(Light Weight Identification,LWID)。该方法首先通过将原始时序数据进行图片化重构,最大限度地保留了不同载波间的特征信息;然后通过设计一种仿生的Balloon机制,实现了对网络层中神经元数量的裁剪,并通过联合使用不同尺寸的卷积核,实现了对数据中特征的提取与特征图中通道信息的整合,从而在提高模型分类能力的前提下实现了模型规模的轻量化。实验结果表明,所提模型在50人的数据集中取得了98.8%的识别率。与传统的基于WiFi信号的身份识别模型相比,所提模型具有更强的分类能力与鲁棒性,同时该模型可以压缩至现有同等精度图片识别模型大小的6.14%。

关键词: LWID, 步态识别, 模型压缩, 频率能量图, Balloon机制

Abstract: As an important research of pervasive computing and human-computer interaction,identity recognition is widely researched.Although traditional WiFi based identification methods have made good progress,they still face challenges such as limi-ted classification ability,high storage cost and long training time.The above problems motivate us to propose a lightweight gait recognition model based on multi-layer neural networks,which is named as LWID(LightWeight Identification).We firstly reconstruct original time series data into graphs to retain characteristic information among different carriers to the maximum extent.Then we design a bionic Balloon mechanism to tailor neurons in network layer.By combining convolution kernels of different size,we extract data features and integrate channel information in the feature map.The proposed method realizes model scale lightweight with higher classification ability.Experimental results show that the model has 98.8% recognition rate in a 50-person dataset.Compared with traditional WiFi based identification model,LWID has stronger classification ability and robustness.Meanwhile,the model is compressed to 6.14% of current computer vision model size with same accuracy.

Key words: Light weight identification, Gait recognition, Model compression, Frequency energy diagram, Balloon mechanism

中图分类号: 

  • TP391
[1] CONNOR P,ROSS A.Biometric recognition by gait:A surveyof modalities and features [J].Computer Vision and Image Understanding,2018,167:1-27.
[2] WU Z F,HUANG Y Z,WANG L,et al.A comprehensive study on cross view gait based hu-man identification with Deep CNNs [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(2):209-226.
[3] ZHANG Y T,PAN G,JIA K,et al.Accelerometer-based gaitrecognition by sparse representation of signature points with clusters [J].IEEE Transactions on Cybernetics,2015,45(9):1864-1875.
[4] AL-NAIMI I,WONG C B,MOORE P,et al.Multi-modal approach for non-tagged indoor identification a-nd tracking using smart floor and pyroelectric infrared sensors [J].International Journal of Computational Science and Engineering,2017,14(1):1-15.
[5] WANG J,JIANGH,XIONG J,et al.LiFS:Low Human-Effort,Device Free Localization with)Fine-Grained)Subcarrier)Information[C]//Proceedings of the 22nd)Annual International Conference on Mobile Computing)and)Networking,2016:243-256.
[6] HAN C,WU K,WANG Y,et al.WiFall:Device-Free Fall Detection by Wireless Networks [C]//Proceedings of IEEE Conference on Computer Communications.2014:271-279.
[7] LI H,YANG W,WANG J,et al.WiFinger:Talk to Your Smart Devices With Finger Grained PH Gesture [C]//Proceedings of the 2016 ACM International JointConference on Pervasive and Ubiquitous Computing,2016:250-261.
[8] LI S J,LI X,NIU K,et al.AR-Alarm:An Adaptive and Robust Intrusion Detection System Leveraging Csi From Commodity Wi-Fi[C]//IntemationaI Conference on Smart Homes and HealthTelematics.2017:211-223.
[9] SONG C,LIN F,ZHUANG Y,et al.Cardiac scan:A non-contact and continuous heart-based user authentication system[C]//Proceedings of the 23rd Annual International Conference on Mobile Computing and Networking.2017:315-328.
[10] PATWARI N,WILSON J,ANANTHANARAYANAN S,etal.Monitoring breathing via signal strength in wireless networks[J].IEEE Trans.Mobile Comput.,2014,13(8):1774-1786.
[11] WANG H,ZHANG D Q,MA J Y,et al.Human respiration de-tection with commodity WiFi devices:Do user location and body orientation matter?[C]//Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp'16).ACM,New York,NY,USA,2016:25-36.
[12] JIN K,CHEN S C.Status and Development of Gait Recognition[J].Computer Science,2019,47(S1):30-34.
[13] BRUTTI A,CAVALLARO A.Online cross-modal adaptation for audiovisual person identification with wearable cameras[J].IEEE Transactions on Human-machine Systems,2017,47(1):40-51.
[14] ALAM M R,BENNAMOUN M,TOGNERI R,et al.A jointdeep Boltzmann machine (jDBM) model for person identification using mobile phone data[J].IEEE Trans.Multimedia,2017,19(2):317-326.
[15] MATSUMURA R,SHIOMI M,MIYASHITA T,et al.Who isinteracting with me? Identification of an interacting person through playful interaction with a small robot[J].IEEE Transactions on Human-Machine Systems,2014,44(2):169-179.
[16] XIONG J,LI F,LIU J.Fusion of different height pyroelectric infrared sensors for person identification[J].IEEE Sensors Journal,2016,16(2):436-446.
[17] SHI C,LIU J,LIU H,et al.Smart user authentication throughactuation of daily activities leveraging WiFi-enabled IoT[C]//Proc.18th ACMInt.Symp.Mobile Ad Hoc Netw.Comput..Chennai,India,2017:1-10.
[18] WANG W,LIU A X,SHAHZAD M.Gait recognition usingWiFi signals[C]//Proceedings of the 18th ACM International Symposium on Mobile Ad Hoc Networking and Computing.2016:363-373.
[19] ZENG Y,PATHAK P H,MOHAPATRA P.WiWho:wifi-based person identification in smart spaces [C]//Proceedings of the 15th International Conference on Information Processing in Sensor Networks.Piscataway,NJ,USA:IEEE,2016:1-12.
[20] ZHANG J,WEI B,HU W,et al.Wifi-id:human identification using wifi signal [C]//Proceedings of the International Conference on Distributed Computing in Sensor Systems.Piscataway,NJ,USA:IEEE,2016:75-82.
[21] XIN T,GUO B,WANG Z,et al.Freesense:Indoor human identification withWi-Fi signals [C]//Proceedings of the IEEE Global Communications Conference.Piscataway,NJ,USA:IEEE,2016:1-7.
[22] OHARA K,MAEKAWA T,MATSUSHITA Y.Detecting state changes of indoor everydayobjects using WiFi channel state information [J].Proceedings of the ACM on Interactive,Mobile,Wearable and Ubiquitous Technologies,2017,1(3):28.
[1] 李青华, 李翠平, 张静, 陈红, 王绍卿. 深度神经网络压缩综述[J]. 计算机科学, 2019, 46(9): 1-14.
[2] 金堃, 陈少昌. 步态识别现状与发展[J]. 计算机科学, 2019, 46(6A): 30-34.
[3] 曾燕, 陈岳林, 蔡晓东. 一种基于权重哈希化的深度人脸识别算法[J]. 计算机科学, 2019, 46(6): 277-281.
[4] 罗正平,刘延钧,杨天奇. 光流分量分解的步态识别[J]. 计算机科学, 2016, 43(9): 295-300.
[5] 李晶,张菁,倪军. 基于修正步态能量图和视角检测的步态识别方法[J]. 计算机科学, 2016, 43(8): 300-303.
[6] 张向刚,唐海,付常君,石宇亮. 一种基于隐马尔科夫模型的步态识别算法[J]. 计算机科学, 2016, 43(7): 285-289.
[7] 刘志勇,冯国灿,陈伟福. 基于局部二值模式和辨识共同向量的步态识别[J]. 计算机科学, 2013, 40(9): 262-265.
[8] 柴艳妹,夏天,韩文英,李海峰. 步态识别研究进展[J]. 计算机科学, 2012, 39(6): 10-15.
[9] 徐艳群,张 斌. 一种基于光流的多区域分割在步态识别中的应用[J]. 计算机科学, 2012, 39(4): 275-277.
[10] 刘志勇,冯国灿,邹小林. 一种基于静态和动态特征的步态识别新方法[J]. 计算机科学, 2012, 39(4): 261-264.
[11] 柴艳妹,韩文英,刘灿涛,李海峰. 融合理论在步态识别中的应用研究[J]. 计算机科学, 2012, 39(12): 272-277.
[12] 陈昌红,梁继民,赵恒,焦李成. 步态表征和步态融合方法新进展[J]. 计算机科学, 2010, 37(8): 15-20.
[13] 申少昊 夏利民 孙正军. 基于遗传模糊分类的步态识别[J]. 计算机科学, 2009, 36(1): 190-193.
[14] 顾磊 吴慧中 肖亮. 一种基于多区域不变矩的步态识别方法[J]. 计算机科学, 2007, 34(11): 167-169.
[15] 吴清江 许文芳 王青力. 基于PCA和SVM的步态识别[J]. 计算机科学, 2006, 33(12): 162-163.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 史雯隽,武继刚,罗裕春. 针对移动云计算任务迁移的快速高效调度算法[J]. 计算机科学, 2018, 45(4): 94 -99 .
[2] 周燕萍,业巧林. 基于L1-范数距离的最小二乘对支持向量机[J]. 计算机科学, 2018, 45(4): 100 -105 .
[3] 刘博艺,唐湘滟,程杰仁. 基于多生长时期模板匹配的玉米螟识别方法[J]. 计算机科学, 2018, 45(4): 106 -111 .
[4] 杨羽琦,章国安,金喜龙. 车载自组织网络中基于车辆密度的双簇头路由协议[J]. 计算机科学, 2018, 45(4): 126 -130 .
[5] 韩奎奎,谢在鹏,吕鑫. 一种基于改进遗传算法的雾计算任务调度策略[J]. 计算机科学, 2018, 45(4): 137 -142 .
[6] 文俊浩,孙光辉,李顺. 基于用户聚类和移动上下文的矩阵分解推荐算法研究[J]. 计算机科学, 2018, 45(4): 215 -219 .
[7] 侯彦娥,孔云峰,党兰学. 求解多车型校车路径问题的混合集合划分的GRASP算法[J]. 计算机科学, 2018, 45(4): 240 -246 .
[8] 秦克云,林洪. 决策形式背景属性约简的关系[J]. 计算机科学, 2018, 45(4): 257 -259 .
[9] 蔡莉,梁宇,朱扬勇,何婧. 数据质量的历史沿革和发展趋势[J]. 计算机科学, 2018, 45(4): 1 -10 .
[10] 梁俊斌,周翔,王田,李陶深. 移动低占空比无线传感网中数据收集的研究进展[J]. 计算机科学, 2018, 45(4): 19 -24 .