Efficient Machine Learning for Wi-Fi CSI-based Human Activity Recognition Using Fast Monte Carlo based Feature Extraction
DOI:
https://doi.org/10.4108/eettti.9442Keywords:
wifi-sensing, doppler, FKV, CSI, SVDAbstract
High-dimensional doppler data extracted from Wi-Fi channel state information (CSI) offers distinctive velocity and time patterns that are useful for human activity recognition (HAR), but its scale poses significant challenges for real-time inference and deployment on resource-constrained devices. This work proposes an efficient, fast monte carlo (MC) feature selection framework based on the frieze-kannanvempala (FKV) algorithm and coefficient estimation to address this bottleneck. The CSI is preprocessed, and doppler traces are computed to encode the velocity and direction of distinct activities. Afterwards, we perform FKV to decompose the doppler data, and the coefficient of the resulting singular vectors is estimated. Using rejection sampling, the topmost features are selected on the basis of their weights, thereby reducing the size of our features. The method identifies a compact set of velocity-time features that preserve critical motion information while significantly reducing computational overhead. The experimental evaluations demonstrated that the decision tree classifier achieved the highest precision at 99.8%, followed by convolutional neural networks (CNN) 96%, the hybrid CNN-long-short-term memory (CNN-LSTM) achieved 87%, while the LSTM model lagged at 53%. These results demonstrated that the integration of fast MC-based feature selection significantly reduced computational overhead without sacrificing classification performance, making it suitable for scalable and real-time HAR applications.
References
[1] M. J. Bocus, K. Chetty, and R. J. Piechocki, “Uwb and wifi sys. as passive opportunistic activity sensing radars,” in 2021 IEEE Radar Conf.(RadarConf21), Atlanta, GA., May 2021, pp. 1–6.
[2] S. P and S. K. Udgata, “Wi-fi sensing enabled violent activity detection in a smart home,” in 2023 IEEE11th Region 10 Humanitarian Technol. Conf. (R10-HTC), Rajkot, India, Oct. 2023, pp. 1274–1279.
[3] G. Apicella, G. D’Aniello, G. Fortino, M. Gaeta, R. Gravina, and L. G. Tramuto, “An adaptive neurofuzzy approach for activity recognition in situation aware wearable systems,” in 2022 IEEE 3rd Int. Conf. Hum.-Mach. Syst. (ICHMS), Orlando, FL, Nov 2022, pp.1–6.
[4] F. Meneghello, C. Chen, C. Cordeiro, and F. Restuccia, “Toward integrated sensing and communications in ieee 802.11bf wi-fi networks,” IEEE Commun. Mag., vol. 61, no. 7, pp. 128–133, Jul 2023.
[5] Y. Dong, X. Cao, Z. Gao, Z. Sun, S. Savovic, X. Zou, W. Pan, L. Yan, and X. Deng, “Resource allocation in multi-user mimo-ofdma vlc systems over low-pass led channels,” IEEE Photon. Technol. Lett., pp. 1–1, Feb 2025.
[6] Y. Golubev, S. Tutelian, V. Loginov, and E. Khorov, “Reducing duration of the mu-mimo csi acquisition procedure in wi-fi networks with ofdma,” in 2023 IEEE Conf. Stand. Commun. Netw. (CSCN), Munich, Germany, 06-08 Nov 2023, pp. 395–400.
[7] C.-F. Hsieh, Y.-C. Chen, C.-Y. Hsieh, and M.-L. Ku, “Device-free indoor human activity recognition using wi-fi rssi: Machine learning approaches,” in 2020 IEEE Int. Conf. Consum. Electron.- Taiwan (ICCE-Taiwan), Taoyuan, Taiwan, Sep 2020, pp. 1–2.
[8] F. Meneghello, D. Garlisi, N. D. Fabbro, I. Tinnirello, and M. Rossi, “SHARP: Environment and person independent activity recognition with commodity IEEE802.11 access points,” IEEE Trans. Mob. Comput., vol. 22, no. 10, pp. 6160–6175, Jun. 2023.
[9] B. Tan, Q. Chen, K. Chetty, K. Woodbridge, W. Li, and R. Piechocki, “ExploitingWiFi channel state information for residential healthcare informatics,” IEEE Commun. Mag., vol. 56, no. 5, pp. 130–137, Aug. 2018.
[10] J. Yang, H. Zou, H. Jiang, and L. Xie, “Device-free occupant activity sensing using wifi-enabled iot devices for smart homes,” IEEE Internet of Things J., vol. 5, no. 5, pp. 3991–4002, June 2018.
[11] I. Nirmal, A. Khamis, M. Hassan, W. Hu, R. Li, and A. Kalyanaraman, “Wifi2radar: Orientationindependent single-receiver wifi sensing via wifi to radar translation,” IEEE Internet of Things J., vol. 11, no. 9, pp. 15 750–15 766, Jan. 2024.
[12] J. Pegoraro, F. Meneghello, and M. Rossi, “Multiperson continuous tracking and identification from mm-wave micro-doppler signatures,” IEEE Trans. Geosci. Remote Sens., vol. 59, no. 4, pp. 2994–3009, Sep 2021.
[13] M. Kim, D. Han, and J.-K. K. Rhee, “Unsupervised view selective deep learning for practical indoor localization using csi,” IEEE Sens. J., vol. 21, no. 21, pp. 24 398–24 408, Sep. 2021.
[14] P. Hu, W. Liu, C. Yang, Y. Sun, and J. Li, “Wifi csi based passive human activity recognition method using blstmcnn,” in 2023 IEEE 23rd Int. Conf. on Commun. Technol. (ICCT), Wuxi, China, Oct. 2023, pp. 210–215.
[15] Z. Shi, Q. Cheng, J. A. Zhang, and R. Yi Da Xu, “Environment-robust wifi-based human activity recognition using enhanced csi and deep learning,” IEEE Internet of Things J., vol. 9, no. 24, pp. 24 643–24 654, Jul. 2022.
[16] M. H. Kabir, M. H. Rahman, and W. Shin, “Csiianet: An inception attention network for human-human interaction recognition based on csi signal,” IEEE Access, vol. 9, pp. 166 624–166 638, Dec 2021.
[17] P. Chen, F. Liu, S. Gao, P. Li, X. Yang, and Q. Niu, “Smartphone-based indoor fingerprinting localization using channel state information,” IEEE Access, vol. 7, pp.180 609–180 619, Dec 2019.
[18] M. M, T. R, V. C, and A. Kanavalli, “Serial vs parallel execution of principal component analysis using singular value decomposition,” in 2024 Int. Conf. Intell. Innov. Technol. Comput., Electr. Electron. (IITCEE), Bangalore, India, 24-25 Jan. 2024, pp. 1–5.
[19] S. Zhang, F. Qin, J. Xu, and J. Zhang, “Small sample scene cancer classification method based on combined feature dimensionality reduction,” in 2024 16th International Conference on Computer and Automation Engineering (ICCAE), Melbourne, Australia, 14-16 Mar. 2024, pp.346–350.
[20] X. Ding,W. Yu, Y. Xie, and S. Liu, “Efficient model-based collaborative filtering with fast adaptive pca,” in 2020IEEE 32nd Int. Conf. Tools Artif. Intell. (ICTAI), Baltimore, MD, 9-11 Nov. 2020, pp. 955–960.
[21] YongchangWang and L. Zhu, “Research and implementation of svd in machine learning,” in 2017 IEEE/ACIS 16th Int. Conf. Comput. Inf. Sci. (ICIS), Kunming, China, 22-24 May 2017, pp. 471–475.
[22] K. J and K. N. Meera, “A hybrid approach for medical image fusion using svd and pca,” in 2024 4th Int. Conf. Innov. Pract. Technol. Manag. (ICIPTM), Noida, India, 21-23 Feb. 2024, pp. 1–5.
[23] D. Janeković and D. Bojanjac, “Randomized algorithms for singular value decomposition: Implementation and application perspective,” in 2021 Int. Symp. ELMAR, Zadar, Croatia, 13-15 Sep. 2021, pp. 165–168.
[24] P. Drineas, R. Kannan, and M. W. Mahoney, “Fast monte carlo algorithms for matrices ii: Computing a low-rank approximation to a matrix,” SIAM J. Comput., vol. 36, no. 1, pp. 158–183, Jan 2006.
[25] H. Cao, X. Shang, Y. Wang, M. Song, S. Chen, and C.-I.Chang, “Go decomposition (GoDec) approach to finding low rank and sparsity matrices for hyperspectral target detection,” in IGARSS 2020 - 2020 IEEE Int. Geosci. Remote Sens. Symp., Waikoloa, HI, 26 Sep. - 02 Oct 2020, pp. 2807–2810.
[26] J. M. Arrazola, A. Delgado, B. R. Bardhan, and S. Lloyd, “Quantum-inspired algorithms in practice,” Quantum, vol. 4, p. 307, Aug. 2020.
[27] A. Gilyén, S. Lloyd, and E. Tang, “Quantum-inspired low-rank stochastic regression with logarithmic dependence on the dimension,” Nov. 2018.
[28] N.-H. Chia, T. Li, H.-H. Lin, and C. Wang, “Quantum inspired sublinear algorithm for solving low-rank semidefinite programming,” jun 2020.
[29] R. Xiao, Z. Zong, and L. Yang, “Clutter suppressionbased on singular value decomposition and fast wavelet algorithm,” in IGARSS 2024 - 2024 IEEE InternationalGeoscience and Remote Sensing Symposium, Athens, Greece, 12-14 Mar. 2024.
[30] Y. Yan, G. Wu, Y. Dong, and Y. Bai, “Floating small target detection in sea clutter based on the singular value decomposition of low rank perturbed random matrices,” in 2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China, 12-14 Mar. 2021, pp. 527–531.
[31] S. Rath and J. Ramalingam, “Comments on “outsourcing eigen-decomposition and singular value decompositionof large matrix to a public cloud”,” IEEE Access, vol. 12, pp. 116 013–116 016, Aug. 2024.
[32] A. Frieze, R. Kannan, and S. Vempala, “Fast montecarlo algorithms for finding low-rank approximations,” in Proc. 39th Annu. Symp. Found. Comput. Sci. (Cat.No.98CB36280), Palo Alto, CA, 08-11 Nov 1998, pp.370–378.15
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Emelia Logah
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
