{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T09:02:03Z","timestamp":1789635723123,"version":"4.0.0"},"reference-count":44,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T00:00:00Z","timestamp":1789603200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key R&D Program of China","doi-asserted-by":"publisher","award":["2024YFC3810805"],"award-info":[{"award-number":["2024YFC3810805"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Hubei Luojia Laboratory","award":["260100022"],"award-info":[{"award-number":["260100022"]}]},{"name":"Hubei Luojia Laboratory","award":["260100025"],"award-info":[{"award-number":["260100025"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Deep learning has achieved notable success in sensor-based human activity recognition (SHAR), yet wearable inertial signals contain activity patterns at different temporal scales while practical deployment imposes strict computational constraints. This paper proposes a three-stage Adaptive CNN-Enhanced Scale-fusion Network (ACESNet) for six-channel accelerometer\u2013gyroscope activity recognition. The Adaptive Kernel Encoder dynamically combines temporal kernels of different sizes, while the CNN-Enhanced Scale-fusion design combines multi-scale temporal mixing, explicit channel mixing and a parallel local CNN path without materializing a global attention matrix. The main benchmark uses a fixed stratified 70\/15\/15 window-level partition (split seed 42), and all methods in the main comparison are repeated over five training seeds. On REALWORLD, MotionSense, UCI-HAR and SHL, ACESNet achieves accuracies of 95.37 \u00b1 0.09%, 99.33 \u00b1 0.12%, 98.01 \u00b1 0.23% and 93.49 \u00b1 0.10%, with Macro-F1 scores of 95.59 \u00b1 0.09%, 99.12 \u00b1 0.15%, 98.16 \u00b1 0.21% and 94.03 \u00b1 0.08%, respectively. A separate official subject-independent UCI-HAR evaluation yields 94.01 \u00b1 0.84% accuracy and 94.11 \u00b1 0.88% Macro-F1. The adaptive-kernel selector has a mean normalized entropy of 0.9114, providing evidence against severe single-branch collapse rather than strong kernel selectivity. On a Raspberry Pi 5 (Raspberry Pi Ltd., Cambridge, UK), ACESNet requires 8.731 MFLOPs and 1.606 \u00b1 0.014 ms per window in FP32 ONNX Runtime 1.18.1. These results support a favorable performance\u2013efficiency trade-off relative to the selected baselines under the stated protocol.<\/jats:p>","DOI":"10.3390\/s26185879","type":"journal-article","created":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T08:03:17Z","timestamp":1789632197000},"page":"5879","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Adaptive Convolutional Neural Network-Enhanced Scale-Fusion Network for Human Activity Recognition Using Wearable Sensors"],"prefix":"10.3390","volume":"26","author":[{"given":"Dongpeng","family":"Xie","sequence":"first","affiliation":[{"name":"GNSS Research Center, Wuhan University, Wuhan 430079, China"},{"name":"Hubei Luojia Laboratory, Wuhan 430079, China"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"methodology","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9661-3562","authenticated-orcid":false,"given":"Peihui","family":"Yan","sequence":"additional","affiliation":[{"name":"GNSS Research Center, Wuhan University, Wuhan 430079, China"},{"name":"Hubei Luojia Laboratory, Wuhan 430079, China"},{"name":"University Student Engineering Training and Innovation Practice Center, Wuhan University, Wuhan 430072, China"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"},{"role":"funding-acquisition","vocabulary":"credit"}]},{"given":"Yifei","family":"Li","sequence":"additional","affiliation":[{"name":"GNSS Research Center, Wuhan University, Wuhan 430079, China"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"software","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"}]},{"given":"Qinghai","family":"Wang","sequence":"additional","affiliation":[{"name":"Hubei Luojia Laboratory, Wuhan 430079, China"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"validation","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"}]},{"given":"Jingnan","family":"Liu","sequence":"additional","affiliation":[{"name":"GNSS Research Center, Wuhan University, Wuhan 430079, China"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"validation","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,17]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"6429","DOI":"10.1109\/JIOT.2020.2985082","article-title":"Deep-Learning-Enhanced Human Activity Recognition for Internet of Healthcare Things","volume":"7","author":"Zhou","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1391906","DOI":"10.1155\/2022\/1391906","article-title":"Wearable sensor-based human activity recognition in the smart healthcare system","volume":"2022","author":"Serpush","year":"2022","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"2820","DOI":"10.1109\/TIM.2018.2829488","article-title":"Improving Photoplethysmographic Measurements Under Motion Artifacts Using Artificial Neural Network for Personal Healthcare","volume":"67","author":"Gupta","year":"2018","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"18499","DOI":"10.1007\/s11042-024-19711-w","article-title":"A review of video-based human activity recognition: Theory, methods and applications","volume":"84","author":"Bukht","year":"2025","journal-title":"Multimed. Tools Appl."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"195719","DOI":"10.1109\/ACCESS.2025.3631392","article-title":"Enhanced Human Activity Recognition Using Mixture of Swin Transformers With Recurrence Plots of Triaxial Sensor Data","volume":"13","author":"Ayub","year":"2025","journal-title":"IEEE Access"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2509313","DOI":"10.1109\/TIM.2023.3265128","article-title":"An Efficient Diverse-Branch Convolution Scheme for Sensor-Based Human Activity Recognition","volume":"72","author":"Han","year":"2023","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"4095","DOI":"10.1007\/s00371-021-02283-3","article-title":"A multibranch CNN-BiLSTM model for human activity recognition using wearable sensor data","volume":"38","author":"Challa","year":"2022","journal-title":"Vis. Comput."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"503291","DOI":"10.1155\/2014\/503291","article-title":"Activity recognition on smartphones via sensor-fusion and KDA-based SVMs","volume":"10","author":"Khan","year":"2014","journal-title":"Int. J. Distrib. Sens. Netw."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Paramasivam, K., Sindha, M.M.R., and Balakrishnan, S.B. (2023). KNN-based machine learning classifier used on deep learned spatial motion features for human action recognition. Entropy, 25.","DOI":"10.3390\/e25060844"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Mekruksavanich, S., and Jitpattanakul, A. (2021). Lstm networks using smartphone data for sensor-based human activity recognition in smart homes. Sensors, 21.","DOI":"10.3390\/s21051636"},{"key":"ref_11","first-page":"1832298","article-title":"STA-HAR: A Spatiotemporal Attention-Based Framework for Human Activity Recognition","volume":"2024","author":"Khaliluzzaman","year":"2024","journal-title":"Appl. Comput. Intell. Soft Comput."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2700316","DOI":"10.1109\/JTEHM.2022.3177710","article-title":"Deep CNN-LSTM With Self-Attention Model for Human Activity Recognition Using Wearable Sensor","volume":"10","author":"Khatun","year":"2022","journal-title":"IEEE J. Transl. Eng. Health Med."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Ord\u00f3\u00f1ez, F.J., and Roggen, D. (2016). Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition. Sensors, 16.","DOI":"10.3390\/s16010115"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1016\/j.eswa.2016.04.032","article-title":"Human activity recognition with smartphone sensors using deep learning neural networks","volume":"59","author":"Ronao","year":"2016","journal-title":"Expert Syst. Appl."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2512313","DOI":"10.1109\/TIM.2021.3102735","article-title":"Deep Neural Networks for Sensor-Based Human Activity Recognition Using Selective Kernel Convolution","volume":"70","author":"Gao","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_16","first-page":"2516912","article-title":"Temporal\u2013Spatial Dynamic Convolutional Neural Network for Human Activity Recognition Using Wearable Sensors","volume":"72","author":"Li","year":"2023","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Dirgov\u00e1 Lupt\u00e1kov\u00e1, I., Kubov\u010d\u00edk, M., and Posp\u00edchal, J. (2022). Wearable sensor-based human activity recognition with transformer model. Sensors, 22.","DOI":"10.20944\/preprints202202.0111.v1"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Maaz, M., Shaker, A., Cholakkal, H., Khan, S., Zamir, S.W., Anwer, R.M., and Shahbaz Khan, F. (2022). Edgenext: Efficiently amalgamated cnn-transformer architecture for mobile vision applications. Proceedings of the European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-031-25082-8_1"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Bibb\u00f2, L., Lagan\u00e0, F., Pullano, S.A., and Angiulli, G. (2026). Multimodal EEG\u2013EMG and FEM-Based Adaptive Control of Passive Upper-Limb Exoskeletons. Sensors, 26.","DOI":"10.3390\/s26123924"},{"key":"ref_20","first-page":"177","article-title":"A survey on deep learning for human activity recognition","volume":"54","author":"Gu","year":"2021","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/j.patrec.2018.02.010","article-title":"Deep learning for sensor-based activity recognition: A survey","volume":"119","author":"Wang","year":"2019","journal-title":"Pattern Recognit. Lett."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"7598","DOI":"10.1109\/JSEN.2019.2917225","article-title":"Attention-based convolutional neural network for weakly labeled human activities\u2019 recognition with wearable sensors","volume":"19","author":"Wang","year":"2019","journal-title":"IEEE Sens. J."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2106","DOI":"10.1109\/TIE.2022.3161812","article-title":"Multiscale deep feature learning for human activity recognition using wearable sensors","volume":"70","author":"Tang","year":"2022","journal-title":"IEEE Trans. Ind. Electron."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1662","DOI":"10.1109\/ACCESS.2017.2779939","article-title":"LSTM fully convolutional networks for time series classification","volume":"6","author":"Karim","year":"2017","journal-title":"IEEE Access"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Li, Y., and Wang, L. (2022). Human activity recognition based on residual network and BiLSTM. Sensors, 22.","DOI":"10.3390\/s22020635"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"56855","DOI":"10.1109\/ACCESS.2020.2982225","article-title":"LSTM-CNN architecture for human activity recognition","volume":"8","author":"Xia","year":"2020","journal-title":"IEEE Access"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"35394","DOI":"10.1109\/JSEN.2024.3450499","article-title":"MHCNLS-HAR: Multi-headed CNN-LSTM based human activity recognition leveraging a novel wearable edge device for elderly health care","volume":"24","author":"Gaud","year":"2024","journal-title":"IEEE Sens. J."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Zhou, Y., Zhao, H., Huang, Y., Riedel, T., Hefenbrock, M., and Beigl, M. (2022, January 11\u201315). Tinyhar: A lightweight deep learning model designed for human activity recognition. Proceedings of the 2022 ACM International Symposium on Wearable Computers, Cambridge, UK.","DOI":"10.1145\/3544794.3558467"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1717","DOI":"10.1109\/JSEN.2023.3335213","article-title":"An improved deep convolutional LSTM for human activity recognition using wearable sensors","volume":"24","author":"Zhang","year":"2023","journal-title":"IEEE Sens. J."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"7878","DOI":"10.1038\/s41598-022-11880-8","article-title":"A Novel CNN-based Bi-LSTM parallel model with attention mechanism for human activity recognition with noisy data","volume":"12","author":"Yin","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"13686","DOI":"10.1109\/JSEN.2024.3373444","article-title":"STFNet: Enhanced and lightweight spatiotemporal fusion network for wearable human activity recognition","volume":"24","author":"Zou","year":"2024","journal-title":"IEEE Sens. J."},{"key":"ref_32","unstructured":"Zhang, X., Song, Y., Song, T., Yang, D., Ye, Y., Zhou, J., and Zhang, L. (2023). AKConv: Convolutional kernel with arbitrary sampled shapes and arbitrary number of parameters. arXiv."},{"key":"ref_33","first-page":"1307","article-title":"Condconv: Conditionally parameterized convolutions for efficient inference","volume":"32","author":"Yang","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., and Liu, Z. (2020, January 13\u201319). Dynamic convolution: Attention over convolution kernels. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01104"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"2328549","DOI":"10.1080\/21642583.2024.2328549","article-title":"Lightweight human activity recognition method based on the MobileHARC model","volume":"12","author":"Gong","year":"2024","journal-title":"Syst. Sci. Control Eng."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Hu, J., Shen, L., and Sun, G. (2018, January 18\u201322). Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Sztyler, T., and Stuckenschmidt, H. (2016, January 14\u201319). On-body localization of wearable devices: An investigation of position-aware activity recognition. Proceedings of the 2016 IEEE International Conference on Pervasive Computing and Communications (PerCom), Sydney, Australia.","DOI":"10.1109\/PERCOM.2016.7456521"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Malekzadeh, M., Clegg, R.G., Cavallaro, A., and Haddadi, H. (2019, January 15\u201318). Mobile sensor data anonymization. Proceedings of the International Conference on Internet of Things Design and Implementation, Montreal, QC, Canada.","DOI":"10.1145\/3302505.3310068"},{"key":"ref_39","unstructured":"Anguita, D., Ghio, A., Oneto, L., Parra, X., and Reyes-Ortiz, J.L. (2013, January 24\u201326). A public domain dataset for human activity recognition using smartphones. Proceedings of the Esann, 2013 Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgium."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"42592","DOI":"10.1109\/ACCESS.2018.2858933","article-title":"The university of sussex-huawei locomotion and transportation dataset for multimodal analytics with mobile devices","volume":"6","author":"Gjoreski","year":"2018","journal-title":"IEEE Access"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"2510811","DOI":"10.1109\/TIM.2021.3091990","article-title":"Shallow Convolutional Neural Networks for Human Activity Recognition Using Wearable Sensors","volume":"70","author":"Huang","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"7414","DOI":"10.1038\/s41598-024-57912-3","article-title":"TCN-attention-HAR: Human activity recognition based on attention mechanism time convolutional network","volume":"14","author":"Wei","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"107728","DOI":"10.1016\/j.asoc.2021.107728","article-title":"DanHAR: Dual Attention Network for multimodal human activity recognition using wearable sensors","volume":"111","author":"Gao","year":"2021","journal-title":"Appl. Soft Comput."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"110143","DOI":"10.1016\/j.knosys.2022.110143","article-title":"TASKED: Transformer-based Adversarial learning for human activity recognition using wearable sensors via Self-KnowledgE Distillation","volume":"260","author":"Suh","year":"2023","journal-title":"Knowl.-Based Syst."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5879\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T08:07:55Z","timestamp":1789632475000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5879"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,17]]},"references-count":44,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26185879"],"URL":"https:\/\/doi.org\/10.3390\/s26185879","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,17]]}}}