{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T17:48:21Z","timestamp":1772300901431,"version":"3.50.1"},"reference-count":14,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2025,7,14]],"date-time":"2025-07-14T00:00:00Z","timestamp":1752451200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Internet Technology Letters"],"published-print":{"date-parts":[[2025,9]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>In this article, we propose a novel TinyML\u2010based framework for real\u2010time sports command recognition under mobile conditions. Unlike conventional Human Activity Recognition (HAR) systems that rely on cloud\u2010based processing or heavy on\u2010device models, our method leverages lightweight deep neural networks, personalized transfer learning, and signal augmentation techniques to perform low\u2010latency and energy\u2010efficient inference directly on microcontroller\u2010class devices. The system is designed to recognize a set of critical sports instructions (e.g., \u201cStart Running,\u201d \u201cJump,\u201d and \u201cSprint\u201d) in mobile or outdoor environments using only wearable inertial sensors. Extensive experiments demonstrate our method outperforms several state\u2010of\u2010the\u2010art baselines in accuracy (95.8%), model size (14.5\u2009KB), and energy efficiency (0.82\u2009mJ per inference). Compared to prior wearable HAR systems, our method uniquely integrates motion\u2010aware segmentation and user\u2010personalized few\u2010shot adaptation, resulting in a 5.3% accuracy gain and 4\u00d7 model compression over baseline TinyML frameworks. The proposed method provides an effective balance between model accuracy, generalization, and hardware efficiency, even in scenarios with significant motion noise and environmental variability.<\/jats:p>","DOI":"10.1002\/itl2.70090","type":"journal-article","created":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T05:04:51Z","timestamp":1752555891000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["<scp>TinyML<\/scp>\n                    \u2010Driven On\u2010Device Sports Command Recognition in Mobile and Dynamic Environments"],"prefix":"10.1002","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-0686-3849","authenticated-orcid":false,"given":"Jiali","family":"Zang","sequence":"first","affiliation":[{"name":"Qiqihar University  Qiqihar Heilongjiang China"}]}],"member":"311","published-online":{"date-parts":[[2025,7,14]]},"reference":[{"key":"e_1_2_7_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2018.02.010"},{"key":"e_1_2_7_3_1","doi-asserted-by":"publisher","DOI":"10.32604\/cmc.2024.054895"},{"key":"e_1_2_7_4_1","volume-title":"Tinyml: Machine Learning With Tensorflow Lite on Arduino and Ultra\u2010Low\u2010Power Microcontrollers","author":"Warden P.","year":"2019"},{"key":"e_1_2_7_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCE.2024.3513331"},{"key":"e_1_2_7_6_1","unstructured":"N. Y.Hammerla S.Halloran andT.Pl\u00f6tz \u201cDeep Convolutional and Recurrent Models for Human Activity Recognition Using Wearables \u201d2016 arXiv preprint arXiv:1604.08880."},{"key":"e_1_2_7_7_1","first-page":"1","volume-title":"23th International Conference on Architecture of Computing Systems","author":"Avci A.","year":"2010"},{"key":"e_1_2_7_8_1","unstructured":"D.BurnsandC.Whyne \u201cPersonalized Activity Recognition With Deep Triplet Embeddings \u201d2020 arXiv preprint arXiv:2001.05517."},{"key":"e_1_2_7_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/1964897.1964918"},{"key":"e_1_2_7_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11036-019-01445-x"},{"key":"e_1_2_7_11_1","doi-asserted-by":"publisher","DOI":"10.3390\/s21041264"},{"key":"e_1_2_7_12_1","first-page":"31","volume-title":"Advances in Neural Information Processing Systems","author":"Kusupati A.","year":"2018"},{"key":"e_1_2_7_13_1","doi-asserted-by":"publisher","DOI":"10.3390\/s21030885"},{"key":"e_1_2_7_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2024.109629"},{"key":"e_1_2_7_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2021.108422"}],"container-title":["Internet Technology Letters"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/itl2.70090","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,21]],"date-time":"2025-10-21T18:26:18Z","timestamp":1761071178000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/itl2.70090"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,14]]},"references-count":14,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,9]]}},"alternative-id":["10.1002\/itl2.70090"],"URL":"https:\/\/doi.org\/10.1002\/itl2.70090","archive":["Portico"],"relation":{},"ISSN":["2476-1508","2476-1508"],"issn-type":[{"value":"2476-1508","type":"print"},{"value":"2476-1508","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,14]]},"assertion":[{"value":"2025-05-25","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-01","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-14","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70090"}}