{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T09:53:57Z","timestamp":1781603637521,"version":"3.54.5"},"reference-count":40,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Regional Innovation System and Education (RISE) program through the Ulsan RISE Center","award":["2026-RISE-07-001"],"award-info":[{"award-number":["2026-RISE-07-001"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>WiFi channel state information (CSI) has become a compelling sensing modality for contactless human activity recognition. However, differences in datasets, preprocessing protocols and model configurations make consistent comparison and reproducibility challenging. This study presents a unified baseline evaluation of four widely adopted deep learning architectures: multilayer perceptron (MLP), convolutional neural network (CNN), gated recurrent unit (GRU) and a hybrid CNN\u2013GRU model across multiple publicly available CSI datasets encompassing a range of sensing tasks. We harmonize the datasets, implement a standardized preprocessing and training pipeline to reduce experimental inconsistencies and support controlled within-dataset comparisons of model behavior. Evaluations include single-person activity recognition, fall-risk estimation, multiperson occupancy classification and localization-aware activity recognition, representing progressively higher temporal and spatial complexity. Our results show dataset-dependent trends: CNNs provide an efficient accuracy\u2013complexity trade-off in several structured activity scenarios, whereas GRUs are advantageous when temporal dynamics are more prominent, although with greater training and inference costs. In contrast, MLPs generally underperform due to limited capacity to capture spatial and temporal dependencies. Confusion matrix analysis reveals that dynamic behaviors and low-motion states remain challenging to distinguish, underscoring the importance of temporal modeling. By releasing the complete experimental pipeline and benchmarking results, this work establishes a reproducible reference framework for the research community and highlights directions for future investigation, including cross-dataset generalization, hybrid model design and lightweight deployment strategies.<\/jats:p>","DOI":"10.3390\/s26123821","type":"journal-article","created":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T09:06:52Z","timestamp":1781600812000},"page":"3821","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep-Learning-Based Baseline Evaluation of Public WiFi CSI Datasets for Contactless RF-Based Human Activity Recognition"],"prefix":"10.3390","volume":"26","author":[{"given":"Tayyaba","family":"Parveen","sequence":"first","affiliation":[{"name":"Department of Electrical Electronic and Computer Engineering, University of Ulsan, Ulsan 44610, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-2429-1669","authenticated-orcid":false,"given":"Rehan","family":"Khan","sequence":"additional","affiliation":[{"name":"Department of Electrical Electronic and Computer Engineering, University of Ulsan, Ulsan 44610, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7666-838X","authenticated-orcid":false,"given":"Umer","family":"Saeed","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, University of Bedfordshire, Luton LU1 3JU, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7476-8782","authenticated-orcid":false,"given":"Insoo","family":"Koo","sequence":"additional","affiliation":[{"name":"Department of Electrical Electronic and Computer Engineering, University of Ulsan, Ulsan 44610, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,16]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Kim Geok, T., Zar Aung, K., Sandar Aung, M., Thu Soe, M., Abdaziz, A., Pao Liew, C., Hossain, F., Tso, C.P., and Yong, W.H. (2020). Review of indoor positioning: Radio wave technology. Appl. Sci., 11.","DOI":"10.3390\/app11010279"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"11483","DOI":"10.1109\/JSEN.2026.3671464","article-title":"AI-Enabled Human Activity Recognition: Bridging Contact-Based and RF-Based Contactless Sensing Paradigms\u2014A Review","volume":"26","author":"Parveen","year":"2026","journal-title":"IEEE Sens. J."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"511","DOI":"10.1109\/TMC.2016.2557795","article-title":"RT-Fall: A real-time and contactless fall detection system with commodity WiFi devices","volume":"16","author":"Wang","year":"2016","journal-title":"IEEE Trans. Mob. Comput."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"171","DOI":"10.1109\/RBME.2022.3156810","article-title":"Contactless WiFi sensing and monitoring for future healthcare-emerging trends, challenges, and opportunities","volume":"16","author":"Ge","year":"2022","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Saeed, U., Shah, S.Y., Shah, S.A., Ahmad, J., Alotaibi, A.A., Althobaiti, T., Ramzan, N., Alomainy, A., and Abbasi, Q.H. (2021). Discrete human activity recognition and fall detection by combining FMCW RADAR data of heterogeneous environments for independent assistive living. Electronics, 10.","DOI":"10.3390\/electronics10182237"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Ali, M., Hendriks, P., Popping, N., Levi, S., and Navee, A. (2023). A comparison of machine learning algorithms for wi-fi sensing using (csi) data. Electronics, 12.","DOI":"10.3390\/electronics12183935"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Duan, P., Diao, X., Cao, Y., Zhang, D., Zhang, B., and Kong, J. (2023). A comprehensive survey on Wi-Fi sensing for human identity recognition. Electronics, 12.","DOI":"10.3390\/electronics12234858"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Saeed, U., Yaseen Shah, S., Aziz Shah, S., Liu, H., Alhumaidi Alotaibi, A., Althobaiti, T., Ramzan, N., Ullah Jan, S., Ahmad, J., and Abbasi, Q.H. (2022). Multiple participants\u2019 discrete activity recognition in a well-controlled environment using universal software radio peripheral wireless sensing. Sensors, 22.","DOI":"10.3390\/s22030809"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Murad, A., and Pyun, J.Y. (2017). Deep recurrent neural networks for human activity recognition. Sensors, 17.","DOI":"10.3390\/s17112556"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Khan, R., Saeed, U., and Koo, I. (2024). FedLSTM: A federated learning framework for sensor fault detection in wireless sensor networks. Electronics, 13.","DOI":"10.3390\/electronics13244907"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Lim, G., Oh, B., Kim, D., and Toh, K.A. (2023). Human activity recognition via score level fusion of Wi-Fi CSI signals. Sensors, 23.","DOI":"10.3390\/s23167292"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1118","DOI":"10.1109\/JSAC.2017.2679658","article-title":"Device-free human activity recognition using commercial WiFi devices","volume":"35","author":"Wang","year":"2017","journal-title":"IEEE J. Sel. Areas Commun."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Khan, R., Saeed, U., and Koo, I. (2025). A Federated Learning Framework with GAF-CNNs for Robust Sensor Fault Detection in WSNs. 2025 20th International Conference on Emerging Technologies (ICET), IEEE.","DOI":"10.1109\/ICET66147.2025.11321454"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1016\/j.neucom.2020.02.137","article-title":"Towards CSI-based diversity activity recognition via LSTM-CNN encoder-decoder neural network","volume":"444","author":"Guo","year":"2021","journal-title":"Neurocomputing"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"5993","DOI":"10.1007\/s00521-021-06787-w","article-title":"Utilizing deep learning models in CSI-based human activity recognition","volume":"34","author":"Shalaby","year":"2022","journal-title":"Neural Comput. Appl."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Sharma, L., Chao, C.H., Wu, S.L., and Li, M.C. (2021). High accuracy wifi-based human activity classification system with time-frequency diagram cnn method for different places. Sensors, 21.","DOI":"10.3390\/s21113797"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Zheng, Y., Zhang, Y., Qian, K., Zhang, G., Liu, Y., Wu, C., and Yang, Z. (2019). Zero-effort cross-domain gesture recognition with Wi-Fi. Proceedings of the 17th Annual International Conference on Mobile Systems, Applications, and Services, Association for Computing Machinery.","DOI":"10.1145\/3307334.3326081"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Khan, R., Saeed, U., and Koo, I. (2025). A conditional gan and dual-channel hybrid deep feature framework for robust sensor fault detection in wsns. 2025 International Conference on Electronics, Information, and Communication (ICEIC), IEEE.","DOI":"10.1109\/ICEIC64972.2025.10879605"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"107171","DOI":"10.1016\/j.engappai.2023.107171","article-title":"WiFi-based human activity recognition through wall using deep learning","volume":"127","author":"Abuhoureyah","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"537","DOI":"10.1007\/s44230-023-00047-x","article-title":"CSI-based location independent human activity recognition using deep learning","volume":"3","author":"Abuhoureyah","year":"2023","journal-title":"Hum.-Centric Intell. Syst."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Kang, H., Kim, D., and Toh, K.A. (2025). Human activity recognition through augmented wifi csi signals by lightweight attention-gru. Sensors, 25.","DOI":"10.3390\/s25051547"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"108232","DOI":"10.1016\/j.compbiomed.2024.108232","article-title":"Efficiently improving the Wi-Fi-based human activity recognition, using auditory features, autoencoders, and fine-tuning","volume":"172","author":"Rahdar","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_23","first-page":"763","article-title":"CSI-based fingerprinting for indoor localization: A deep learning approach","volume":"66","author":"Wang","year":"2016","journal-title":"IEEE Trans. Veh. Technol."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Xia, M., Que, S., Liu, N., Wang, Q., and Li, T. (2025). Motion Pattern Recognition via CNN-LSTM-Attention Model Using Array-Based Wi-Fi CSI Sensors in GNSS-Denied Areas. Electronics, 14.","DOI":"10.3390\/electronics14081594"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"100703","DOI":"10.1016\/j.patter.2023.100703","article-title":"SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing","volume":"4","author":"Yang","year":"2023","journal-title":"Patterns"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3705893","article-title":"Wi-Fi sensing techniques for human activity recognition: Brief survey, potential challenges, and research directions","volume":"57","author":"Miao","year":"2025","journal-title":"ACM Comput. Surv."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"2714","DOI":"10.1109\/TMC.2018.2878233","article-title":"WiFi CSI based passive human activity recognition using attention based BLSTM","volume":"18","author":"Chen","year":"2018","journal-title":"IEEE Trans. Mob. Comput."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"7186","DOI":"10.1109\/TCCN.2026.3683215","article-title":"Neural Approaches to Wi-Fi CSI-Based Human Activity Recognition: Architectures and Generalization","volume":"12","author":"He","year":"2026","journal-title":"IEEE Trans. Cogn. Commun. Netw."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Huang, S., Li, K., You, D., Chen, Y., Lin, A., Liu, S., Li, X., and McCann, J.A. (2024). WiMANS: A benchmark dataset for WiFi-based multi-user activity sensing. Computer Vision\u2014ECCV 2024 (ECCV 2024), Springer.","DOI":"10.1007\/978-3-031-72946-1_5"},{"key":"ref_30","unstructured":"Zhu, G., Hu, Y., Gao, W., Wang, W.H., Wang, B., and Liu, K. (2026). CSI-Bench: A large-scale in-the-wild dataset for multi-task WiFi sensing. Adv. Neural Inf. Process. Syst., 38, Available online: https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2025\/file\/f7c68372b3d39e8d7a093eb2edaaad87-Paper-Datasets_and_Benchmarks_Track.pdf."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"245","DOI":"10.1109\/IEEEDATA.2026.3681245","article-title":"Descriptor: The WiFi-Based Human Activity Recognition Under Everyday Life Changes Dataset (WiFi-Wild)","volume":"3","author":"Elkelany","year":"2026","journal-title":"IEEE Data Descr."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Moghaddam, M.G., Warnakulasuriya, N., Pala, G.S.S.P., Nowdeh, A.R., Shirehjini, A.A.N., and Shirmohammadi, S. (2026, March 12). Descriptor: WiFi-Granular Activities in Kitchen Dataset (Wi-GRAK). IEEE Data Descriptions, Available online: https:\/\/ieee-dataport.org\/documents\/wifi-granular-activities-kitchen-dataset-wi-grak.","DOI":"10.1109\/IEEEDATA.2026.3698410"},{"key":"ref_33","unstructured":"Taylor, W., Pinkerton, S., Khan, M.T., Taha, M., Barakat, B., Shawky, M.A., Abbasi, Q.H., Imran, M.A., and Taha, A. (2026, March 12). WiPE-FaLl: Wi-Fi-Based Prediction and Estimation of Fall Likelihood. Available online: https:\/\/researchdata.gla.ac.uk\/1623\/."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"9018","DOI":"10.1109\/JSEN.2022.3160796","article-title":"Non-invasive localization using software-defined radios","volume":"22","author":"Khan","year":"2022","journal-title":"IEEE Sens. J."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"17590","DOI":"10.1038\/s41598-021-96689-7","article-title":"5g-enabled contactless multi-user presence and activity detection for independent assisted living","volume":"11","author":"Ashleibta","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_36","unstructured":"Yang, J., Chen, X., Wang, D., Zou, H., Lu, C.X., Sun, S., and Xie, L. (2022). Deep learning and its applications to WiFi human sensing: A benchmark and a tutorial. arXiv."},{"key":"ref_37","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A Method for Stochastic Optimization. arXiv."},{"key":"ref_38","unstructured":"Hussain, Z., Sheng, M., and Zhang, W.E. (2019). Different approaches for human activity recognition: A survey. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"13912","DOI":"10.1109\/JSEN.2025.3547736","article-title":"Robust sensor fault detection in wireless sensor networks using a hybrid conditional generative adversarial networks and convolutional autoencoder","volume":"25","author":"Khan","year":"2025","journal-title":"IEEE Sens. J."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"2702","DOI":"10.1038\/s41598-024-53069-1","article-title":"Wireless body area sensor networks based human activity recognition using deep learning","volume":"14","author":"Essa","year":"2024","journal-title":"Sci. Rep."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3821\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T09:22:04Z","timestamp":1781601724000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3821"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,16]]},"references-count":40,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["s26123821"],"URL":"https:\/\/doi.org\/10.3390\/s26123821","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,16]]}}}