Abstract
Human activity recognition plays an important role in a variety of daily applications. There has been tremendous work on human activity recognition based on WiFi channel state information (CSI). Although achieving reasonable performance in certain cases, they are yet faced with a major challenge: location dependence. An activity recognition model trained at one location does not perform properly at other locations, because the human location also has significant influence on WiFi signal propagation. In this paper, we aim to solve the location dependence problem of CSI-based human activity recognition and propose a device-free cross location activity recognition (CLAR) method via semi-supervised deep learning. We regard the locations with labeled activity samples as the source domains and the locations with unlabeled activity samples as the target domains. By exploiting pseudo labeling and feature mapping, CLAR trains an activity recognition model working across the source and the target domains as well as the unseen domains which have no training samples. CLAR first extracts the trend component from the activity samples by Singular Spectrum Analysis (SSA), then annotates the unlabeled samples with the pseudo labels through a dual-score multi-classifier labeling model. The activity recognition model is trained using the labeled samples from the source domains and the pseudo-labeled samples from the target domains. Both the labeling and the recognition models are based on Bidirectional Long Short Term Memory (BLSTM). Evaluations in real-world environments demonstrate the effectiveness and generalization of the method CLAR, which performs well for both the source and the target domains, and generalizes well to the unseen domains.
Similar content being viewed by others
References
Arshad S, Feng C, Liu Y, Hu Y, Yu R, Zhou S, Li H (2017) Wi-Chase: a WiFi based human activity recognition system for sensorless environments. In: WoWMoM’2017, pp. 1–6
Chang JY, Lee KY, Wei YL, Lin KCJ, Hsu W (2016) Location-independent WiFi action recognition via vision-based methods. In: MM’2016, pp. 162–166. ACM
Chen X, Li H, Zhou C, Liu X, Wu D, Dudek G (2020) FiDo: Ubiquitous fine-grained WiFi-based localization for unlabelled users via domain adaptation. In: Proceedings of the Web Conference 2020, WWW’20, pp. 23–33
Chen Z, Zhang L, Jiang C, Cao Z, Cui W (2019) WiFi CSI based passive human activity recognition using attention based BLSTM. IEEE Trans Mob Comput 18(11):2714–2724
Feng C, Arshad S, Liu Y (2017) MAIS: Multiple activity identification system using channel state information of WiFi signals. In: International Conference on Wireless Algorithms, Systems, and Applications, pp. 419–432
Gao Q, Wang J, Ma X, Feng X, Wang H (2017) CSI-based device-free wireless localization and activity recognition using radio image features. IEEE Trans Veh Technol 66(11):10346–10356
Gkioxari G, Girshick R, Dollar P, He K (2018) Detecting and recognizing human-object interactions. In: CVPR’2018, pp. 8359–8367
Golyandina N, Nekrutkin V, Zhigljavsky A (2001) Analysis of time series structure: SSA and related techniques. Monographs on Statistics and Applied Probability 90
Guan Y, Plötz T (2017) Ensembles of deep LSTM learners for activity recognition using wearables. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT) 1(2)
Halperin D, Hu W, Sheth A, Wetherall D (2010) Predictable 802.11 packet delivery from wireless channel measurements. In: SIGCOMM’2010, pp. 159–170. ACM
Jiang W, Miao C, Ma F, Yao S, Wang Y, Yuan Y, Xue H, Song C, Ma X, Koutsonikolas D, Xu W, Su L (2018) Towards environment independent device free human activity recognition. In: MobiCom’2018, pp. 289–304. ACM
Khan MI, Jan MA, Muhammad Y, Do DT, ur Rehman A, Mavromoustakis CX, Pallis E (2021) Tracking vital signs of a patient using channel state information and machine learning for a smart healthcare system. Neural Computing and Applications
Li H, Yang W, Wang J, Xu Y, Huang L (2016) WiFinger: Talk to your smart devices with finger-grained gesture. In: UbiComp’2016, pp. 250–261. ACM
Liu J, Liu H, Chen Y, Wang Y, Wang C (2020) Wireless sensing for human activity: a survey. IEEE Commun Surv Tutor 22(3):1629–1645
Liu X, Cao J, Tang S, Wen J, Guo P (2016) Contactless respiration monitoring via off-the-shelf WiFi devices. IEEE Trans Mob Comput 15(10):2466–2479
Ma Y, Zhou G, Wang S (2019) WiFi sensing with channel state information: a survey. ACM Comput Surv 52(3):1–36
Markopoulos P, Zlotnikov S, Ahmad F (2019) Adaptive radar-based human activity recognition with L1-norm linear discriminant analysis. IEEE J Electromagn RF Microw Med Biol 3(2):120–126
Orphomma S, Swangmuang N (2013) Exploiting the wireless RF fading for human activity recognition. In: 10th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, pp. 1–5
Qi F, Li Z, Liang F, Lv H, An Q, Wang J (2016) A novel time-frequency analysis method based on HHT for finer-grained human activity using SFCW radar. In: 2016 Progress in Electromagnetic Research Symposium (PIERS), pp. 2536–2539
Shang J, Wu J (2016) Fine-grained vital signs estimation using commercial Wi-Fi devices. In: 8th Wireless of the Students, by the Students, and for the Students Workshop, pp. 30–32. ACM
Shen S, Wang H, Roy Choudhury R (2016) I am a smartwatch and I can track my user’s arm. In: Mobisys’2016, pp. 85–96
Sheng B, Xiao F, Sha L, Sun L (2020) Deep spatial-temporal model based cross-scene action recognition using commodity WiFi. IEEE Internet Things J 7(4):3592–3601
Sigg S, Shi S, Ji Y (2013) RF-based device-free recognition of simultaneously conducted activities. In: UbiComp’13 Adjunct, pp. 531–540. ACM
Wang F, Gong W, Liu J (2019) On spatial diversity in WiFi-based human activity recognition: a deep learning-based approach. IEEE Internet Things J 6(2):2035–2047
Wang G, Zou Y, Zhou Z, Wu K, Ni L (2016) We can hear you with Wi-Fi! IEEE Trans Mob Comput 15:2907–2920
Wang H, Zhang D, Ma J, Wang Y, Wang Y, Wu D, Gu T, Xie B (2016) Human respiration detection with commodity wifi devices: Do user location and body orientation matter? In: UbiComp’2016, pp. 25–36. ACM
Wang H, Zhang D, Wang Y, Ma J, Wang Y, Li S (2017) RT-Fall: a real-time and contactless fall detection system with commodity WiFi devices. IEEE Trans Mob Comput 16(2):511–526
Wang J, Zhang X, Gao Q, Yue H, Wang H (2017) Device-free wireless localization and activity recognition: a deep learning approach. IEEE Trans Veh Technol 66(7):6258–6267
Wang M, Ni B, Yang X (2017) Recurrent modeling of interaction context for collective activity recognition. In: CVPR’2017, pp. 7408–7416
Wang W, Liu AX, Shahzad M, Ling K, Lu S (2015) Understanding and modeling of WiFi signal based human activity recognition. In: MobiCom’2015, pp. 65–76. ACM
Wang W, Liu AX, Shahzad M, Ling K, Lu S (2017) Device-free human activity recognition using commercial WiFi devices. IEEE J Sel Areas Commun 35(5):1118–1131
Wang X, Yang C, Mao S (2017) PhaseBeat: Exploiting CSI phase data for vital sign monitoring with commodity WiFi devices. In: ICDCS’2017, pp. 1230–1239
Wang X, Yang C, Mao S (2017) TensorBeat: Tensor decomposition for monitoring multi-person breathing beats with commodity WiFi. ACM Trans Intell Syst Technol. Vol. 9
Wu D, Gao R, Zeng Y, Liu J, Wang L, Gu T, Zhang D (2020) Fingerdraw: sub-wavelength level finger motion tracking with wifi signals. Proc ACM Interact Mobile Wearable Ubiquitous Technol IMWUT 4(1):1–27
Xiao C, Han D, Ma Y, Qin Z (2019) CsiGAN: robust channel state information-based activity recognition with GANs. IEEE Internet Things J 6(6):10191–10204
Zhang F, Niu K, Xiong J, Jin B, Gu T, Jiang Y, Zhang D (2019) Towards a diffraction-based sensing approach on human activity recognition. Proc ACM Interact Mobile Wearable Ubiquitous Technol IMWUT 3(1):1–25
Zhao M, Adib F, Katabi D (2018) Emotion recognition using wireless signals. Commun ACM 61:91–100
Zheng Y, Zhang Y, Qian K, Zhang G, Liu Y, Wu C, Yang Z (2019) Zero-effort cross-domain gesture recognition with Wi-Fi. In: MobiSys’2019, pp. 313–325. ACM
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
No conflict of interest exists in the submission.
Ethical approval
The manuscript is approved by all the co-authors for publication in this journal.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Zhou, R., Gong, Z., Tang, K. et al. Device-free cross location activity recognition via semi-supervised deep learning. Neural Comput & Applic 34, 10189–10203 (2022). https://doi.org/10.1007/s00521-022-07085-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-022-07085-9