{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deep-csi-learning-for-gait-biometric-sensing","title":"Deep CSI Learning for Gait Biometric Sensing and Recognition","arxiv_id":"1902.02300","date":"2019-02-06","proceeding":null,"authors":["Kalvik Jakkala","Arupjyoti Bhuya","Zhi Sun","Pu Wang","Zhuo Cheng"],"abstract":"Gait is a person's natural walking style and a complex biological process\nthat is unique to each person. Recently, the channel state information (CSI) of\nWiFi devices have been exploited to capture human gait biometrics for user\nidentification. However, the performance of existing CSI-based gait\nidentification systems is far from satisfactory. They can only achieve limited\nidentification accuracy (maximum $93\\%$) only for a very small group of people\n(i.e., between 2 to 10). To address such challenge, an end-to-end deep CSI\nlearning system is developed, which exploits deep neural networks to\nautomatically learn the salient gait features in CSI data that are\ndiscriminative enough to distinguish different people Firstly, the raw CSI data\nare sanitized through window-based denoising, mean centering and normalization.\nThe sanitized data is then passed to a residual deep convolutional neural\nnetwork (DCNN), which automatically extracts the hierarchical features of\ngait-signatures embedded in the CSI data. Finally, a softmax classifier\nutilizes the extracted features to make the final prediction about the identity\nof the user. In a typical indoor environment, a top-1 accuracy of $97.12 \\pm\n1.13\\%$ is achieved for a dataset of 30 people.","url_abs":"http://arxiv.org/abs/1902.02300v1","url_pdf":"http://arxiv.org/pdf/1902.02300v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deep-csi-learning-for-gait-biometric-sensing","repo_url":"https://github.com/itskalvik/WiFi-user-recognition","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"gait-identification","task_name":"Gait Identification"},{"task_slug":"user-identification","task_name":"User Identification"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}