{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T15:49:13Z","timestamp":1775490553972,"version":"3.50.1"},"reference-count":45,"publisher":"Association for Computing Machinery (ACM)","issue":"2","funder":[{"name":"Japan Science and Technology Agency (JST) PRESTO","award":["JPMJPR21P7"],"award-info":[{"award-number":["JPMJPR21P7"]}]},{"name":"JSPS KAKENHI","award":["JP24K15227"],"award-info":[{"award-number":["JP24K15227"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Comput. Healthcare"],"published-print":{"date-parts":[[2026,4,30]]},"abstract":"<jats:p>Wearable accelerometers and gyroscopes capture fine-grained behavioral signatures that can inadvertently reveal user identities, making privacy protection essential for healthcare applications. We present C-AAE, a lightweight compressive anonymizing autoencoder that performs on-device privacy filtering at the sensor edge. The core idea of C-AAE is to integrate two complementary privacy filters: a learned, sensor-specific anonymization module, the Anonymizing AutoEncoder (AAE), and a learning-free, generic anonymization module, Adaptive Differential Pulse-Code Modulation (ADPCM). The AAE locally learns to suppress identity cues while preserving activity-relevant representations, whereas ADPCM provides training-free anonymization through compression, further masking residual identity information and reducing communication cost. Experiments on the MotionSense and PAMAP2 datasets show that C-AAE cuts user re-identification F1 scores by 10\u201315 percentage points relative to AAE alone, while keeping activity-recognition F1 within 5 percentage points of the unprotected baseline. Implementation on a small-scale edge device (ESP32-WROOM-32) demonstrates real-time performance with markedly lower memory usage, latency, and power consumption. Unlike differential-privacy mechanisms that rely on randomized noise, C-AAE offers a complementary, representation-level approach, enabling practical and resource-efficient on-device anonymization that remains compatible with formal DP frameworks for hybrid deployment on edge healthcare devices.<\/jats:p>","DOI":"10.1145\/3793553","type":"journal-article","created":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T14:21:04Z","timestamp":1770646864000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["C-AAE: A Compressive Anonymizing AutoEncoder for Privacy-Preserving Activity Recognition on Edge Devices"],"prefix":"10.1145","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-1205-5408","authenticated-orcid":false,"given":"Ryusei","family":"Fujimoto","sequence":"first","affiliation":[{"name":"Kyushu University, Fukuoka, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-5266-8352","authenticated-orcid":false,"given":"Musashi","family":"Hadano","sequence":"additional","affiliation":[{"name":"Kyushu University, Fukuoka, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8834-5323","authenticated-orcid":false,"given":"Yugo","family":"Nakamura","sequence":"additional","affiliation":[{"name":"Kyushu University, Fukuoka, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7156-9160","authenticated-orcid":false,"given":"Yutaka","family":"Arakawa","sequence":"additional","affiliation":[{"name":"Kyushu University, Fukuoka, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2026,4,6]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.3390\/make6020065"},{"key":"e_1_3_1_3_2","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1007\/978-3-030-10997-4_33","volume-title":"Machine Learning and Knowledge Discovery in Databases","author":"Bevilacqua Antonio","year":"2019","unstructured":"Antonio Bevilacqua, Kyle MacDonald, Aamina Rangarej, Venessa Widjaya, Brian Caulfield, and Tahar Kechadi. 2019. Human activity recognition with convolutional neural networks. In Machine Learning and Knowledge Discovery in Databases. Ulf Brefeld, Edward Curry, Elizabeth Daly, Brian MacNamee, Alice Marascu, Fabio Pinelli, Michele Berlingerio, and Neil Hurley (Eds.). Springer International Publishing, Cham, 541\u2013552."},{"key":"e_1_3_1_4_2","doi-asserted-by":"crossref","first-page":"101189","DOI":"10.1016\/j.iot.2024.101189","article-title":"Privacy preservation in sensor-based human activity recognition through autoencoders for low-power IoT devices","volume":"26","author":"Bigelli L.","year":"2024","unstructured":"L. Bigelli, C. Contoli, V. Freschi, and E. Lattanzi. 2024. Privacy preservation in sensor-based human activity recognition through autoencoders for low-power IoT devices. Internet of Things 26 (2024) 101189.","journal-title":"Internet of Things"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1023\/A:1010933404324"},{"key":"e_1_3_1_6_2","doi-asserted-by":"crossref","first-page":"792065","DOI":"10.3389\/fcomp.2021.792065","article-title":"Opportunity++: A multimodal dataset for video-and wearable, object and ambient sensors-based human activity recognition","volume":"3","author":"Ciliberto Mathias","year":"2021","unstructured":"Mathias Ciliberto, Vitor Fortes Rey, Alberto Calatroni, Paul Lukowicz, and Daniel Roggen. 2021. Opportunity++: A multimodal dataset for video-and wearable, object and ambient sensors-based human activity recognition. Frontiers in Computer Science 3 (2021), 792065.","journal-title":"Frontiers in Computer Science"},{"key":"e_1_3_1_7_2","doi-asserted-by":"crossref","first-page":"48046","DOI":"10.1109\/ACCESS.2023.3276438","article-title":"A study on the application of TensorFlow compression techniques to human activity recognition","volume":"11","author":"Contoli Chiara","year":"2023","unstructured":"Chiara Contoli and Emanuele Lattanzi. 2023. A study on the application of TensorFlow compression techniques to human activity recognition. IEEE Access 11 (2023), 48046\u201348058.","journal-title":"IEEE Access"},{"issue":"13","key":"e_1_3_1_8_2","doi-asserted-by":"crossref","first-page":"1105","DOI":"10.1002\/j.1538-7305.1973.tb02007.x","article-title":"Adaptive quantization in differential PCM coding of speech","volume":"52","author":"Cummiskey P.","year":"1973","unstructured":"P. Cummiskey, N. S. Jayant, and J. L. Flanagan. 1973. Adaptive quantization in differential PCM coding of speech. Bell System Technical Journal 52, 13 (1973), 1105\u20131118.","journal-title":"Bell System Technical Journal"},{"key":"e_1_3_1_9_2","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","author":"Dem\u0161ar Janez","year":"2006","unstructured":"Janez Dem\u0161ar. 2006. Statistical comparisons of classifiers over multiple data sets. The Journal of Machine Learning Research 7 (2006), 1\u201330.","journal-title":"The Journal of Machine Learning Research"},{"key":"e_1_3_1_10_2","first-page":"1","volume-title":"Automata, Languages and Programming","author":"Cynthia Dwork","year":"2006","unstructured":"Dwork Cynthia. 2006. Differential privacy. In Automata, Languages and Programming. Michele Bugliesi, Bart Preneel, Vladimiro Sassone, and Ingo Wegener (Eds.). Springer, Berlin, 1\u201312."},{"key":"e_1_3_1_11_2","first-page":"1","volume-title":"Privacy, Security, and Trust in KDD","author":"Cynthia Dwork","year":"2008","unstructured":"Dwork Cynthia. 2008. An ad omnia approach to defining and achieving private data analysis. In Privacy, Security, and Trust in KDD. Francesco Bonchi, Elena Ferrari, Bradley Malin, and Y\u00fccel Saygin (Eds.). Springer, Berlin, 1\u201313."},{"issue":"1","key":"e_1_3_1_12_2","first-page":"113","article-title":"LightGBM for human activity recognition using wearable sensors","volume":"5","author":"Fu Pengxi","year":"2024","unstructured":"Pengxi Fu, Jianxin Guo, and Hongxiang Luo. 2024. LightGBM for human activity recognition using wearable sensors. Automation and Machine Learning 5, 1 (2024), 113\u2013118.","journal-title":"Automation and Machine Learning"},{"key":"e_1_3_1_13_2","volume-title":"2023 IEEE International Conference on Pervasive Computing and Communications Workshops and Other Affiliated Events (PerCom Workshops, PrivaCom \u201923)","author":"Fujimoto Ryusei","year":"2023","unstructured":"Ryusei Fujimoto, Yugo Nakamura, and Yutaka Arakawa. 2023. Differential privacy with weighted \\(\\epsilon\\) for privacy-preservation in human activity recognition. In 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and Other Affiliated Events (PerCom Workshops, PrivaCom \u201923)."},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-022-13185-4"},{"key":"e_1_3_1_15_2","volume-title":"Deep Learning","author":"Goodfellow Ian","year":"2016","unstructured":"Ian Goodfellow, Yoshua Bengio, and Aaron Courville. 2016. Deep Learning. MIT Press. Retrieved from http:\/\/www.deeplearningbook.org"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.21105\/joss.02173"},{"key":"e_1_3_1_17_2","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2023.3347552"},{"key":"e_1_3_1_18_2","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1109\/MDM58254.2023.00042","volume-title":"2023 24th IEEE International Conference on Mobile Data Management (MDM)","author":"Hidayat Muhammad Ayat","year":"2023","unstructured":"Muhammad Ayat Hidayat, Yugo Nakamura, Billy Dawton, and Yutaka Arakawa. 2023. AGC-DP: Differential privacy with adaptive Gaussian clipping for federated learning. In 2023 24th IEEE International Conference on Mobile Data Management (MDM), 199\u2013208. DOI: 10.1109\/MDM58254.2023.00042"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3235969"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2019.2909688"},{"key":"e_1_3_1_21_2","first-page":"1401","volume-title":"2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT)","author":"Ismail Iman Khalid","year":"2024","unstructured":"Khalid Ismail Iman, R. I. Boby, Md Sazib Mollik, Imran Chowdhury, Sahidur Rahman, Wobeydul Haq Ankur, and Khaizuran Abdullah. 2024. A real-time IoT cryptograph communication of chaotic-ADPCM coding method for IoT applications. In 2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT), 1401\u20131406. DOI: 10.1109\/ICEEICT62016.2024.10534526"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2016.2633620"},{"key":"e_1_3_1_23_2","doi-asserted-by":"crossref","first-page":"2637","DOI":"10.3390\/s22072637","article-title":"Exploring artificial neural networks efficiency in tiny wearable devices for human activity recognition","volume":"22","author":"Lattanzi E.","year":"2022","unstructured":"E. Lattanzi, M. Donati, and V. Freschi. 2022. Exploring artificial neural networks efficiency in tiny wearable devices for human activity recognition. Sensors 22 (2022), 2637.","journal-title":"Sensors"},{"key":"e_1_3_1_24_2","first-page":"3657282","article-title":"Lightweight deep learning for resource-constrained environments: A survey","author":"Liu H.-I.","year":"2024","unstructured":"H.-I. Liu, M. Galindo, H. Xie, L. K. Wong, H. H. Shuai, Y. H. Li, and W. H. Cheng. 2024. Lightweight deep learning for resource-constrained environments: A survey. Computing Surveys 56 (May. 2024), 3657282.","journal-title":"Computing Surveys"},{"key":"e_1_3_1_25_2","first-page":"49","volume-title":"International Conference on Internet of Things Design and Implementation (IoTDI \u201919)","author":"Malekzadeh Mohammad","year":"2019","unstructured":"Mohammad Malekzadeh, Richard G. Clegg, Andrea Cavallaro, and Hamed Haddadi. 2019. Mobile sensor data anonymization. In International Conference on Internet of Things Design and Implementation (IoTDI \u201919). ACM, New York, NY, 49\u201358. DOI: 10.1145\/3302505.3310068"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.pmcj.2020.101132"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.3390\/electronics10030308"},{"key":"e_1_3_1_28_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jisa.2022.103204"},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/3771551"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/MPRV.2025.3586147"},{"key":"e_1_3_1_31_2","unstructured":"Nicolas Papernot Mart\u00edn Abadi \u00dalfar Erlingsson Ian J. Goodfellow and Kunal Talwar. 2016. Semi-supervised knowledge transfer for deep learning from private training data. arXiv:161005755. Retrieved from https:\/\/arxiv.org\/abs\/1610.05755"},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2022.108778"},{"key":"e_1_3_1_33_2","unstructured":"Alec Radford Luke Metz and Soumith Chintala. 2015. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:1511.06434. Retrieved from https:\/\/arxiv.org\/abs\/1511.06434"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.24432\/C5NW2H"},{"key":"e_1_3_1_35_2","first-page":"1","volume-title":"2023 IEEE 10th International Conference on Data Science and Advanced Analytics (DSAA)","author":"Roy D.","year":"2023","unstructured":"D. Roy and S. Girdzijauskas. 2023. Temporal differential privacy for human activity recognition. In 2023 IEEE 10th International Conference on Data Science and Advanced Analytics (DSAA), 1\u201310. DOI: 10.1109\/DSAA60987"},{"key":"e_1_3_1_36_2","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1016\/j.patrec.2021.02.024","article-title":"KU-HAR: An open dataset for heterogeneous human activity recognition","volume":"146","author":"Sikder Niloy","year":"2021","unstructured":"Niloy Sikder and Abdullah-Al Nahid. 2021. KU-HAR: An open dataset for heterogeneous human activity recognition. Pattern Recognition Letters 146 (2021), 46\u201354.","journal-title":"Pattern Recognition Letters"},{"key":"e_1_3_1_37_2","volume-title":"11th ACM Symposium on Eye Tracking Research & Applications (ETRA \u201919)","author":"Steil Julian","year":"2019","unstructured":"Julian Steil, Inken Hagestedt, Michael Xuelin Huang, and Andreas Bulling. 2019. Privacy-aware eye tracking using differential privacy. In 11th ACM Symposium on Eye Tracking Research & Applications (ETRA \u201919). ACM, New York, NY, Article 27, 1\u20139. DOI: 10.1145\/3314111.3319915"},{"key":"e_1_3_1_38_2","doi-asserted-by":"crossref","first-page":"56571","DOI":"10.1109\/ACCESS.2024.3390039","article-title":"Preserving data utility in differentially private smart home data","author":"Stirapongsasuti Sopicha","year":"2024","unstructured":"Sopicha Stirapongsasuti, Francis Jerome Tiausas, Yugo Nakamura, and Keiichi Yasumoto. 2024. Preserving data utility in differentially private smart home data. IEEE Access (2024), 56571\u201356581.","journal-title":"IEEE Access"},{"issue":"05","key":"e_1_3_1_39_2","doi-asserted-by":"crossref","first-page":"571","DOI":"10.1142\/S021848850200165X","article-title":"Achieving k-anonymity privacy protection using generalization and suppression","volume":"10","author":"Sweeney Latanya","year":"2002","unstructured":"Latanya Sweeney. 2002. Achieving k-anonymity privacy protection using generalization and suppression. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems 10, 05 (2002), 571\u2013588.","journal-title":"International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems"},{"key":"e_1_3_1_40_2","unstructured":"The IMA Digital Audio Focus and Technical Working Groups. 1992. Recommended practices for enhancing digital audio compatibility in multimedia systems. DATWG Recommendation 1\u201340."},{"key":"e_1_3_1_41_2","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1109\/ALLERTON.2019.8919758","volume-title":"2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton)","author":"Tripathy Ardhendu","year":"2019","unstructured":"Ardhendu Tripathy, Ye Wang, and Prakash Ishwar. 2019. Privacy-preserving adversarial networks. In 2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton). IEEE Press, 495\u2013505. DOI: 10.1109\/ALLERTON.2019.8919758"},{"key":"e_1_3_1_42_2","first-page":"143","volume-title":"International Conference on Information and Communication Technologies for Ageing Well and e-Health","author":"Vavoulas George","year":"2016","unstructured":"George Vavoulas, Charikleia Chatzaki, Thodoris Malliotakis, Matthew Pediaditis, and Manolis Tsiknakis. 2016. The mobiact dataset: Recognition of activities of daily living using smartphones. In International Conference on Information and Communication Technologies for Ageing Well and e-Health, Vol. 2. SciTePress, 143\u2013151."},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/LSENS.2017.2726759"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.3390\/s21134602"},{"key":"e_1_3_1_45_2","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1109\/CPHS56133.2022.9804533","volume-title":"2022 2nd International Workshop on Cyber-Physical-Human System Design and Implementation (CPHS)","author":"Zhan Yuting","year":"2022","unstructured":"Yuting Zhan, Hamed Haddadi, Alex Kyllo, and Afra Mashhadi. 2022. Privacy-aware human mobility prediction via adversarial networks. In 2022 2nd International Workshop on Cyber-Physical-Human System Design and Implementation (CPHS), 7\u201312. DOI: 10.1109\/CPHS56133.2022.9804533"},{"key":"e_1_3_1_46_2","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1145\/3675095.3676624","volume-title":"2024 ACM International Symposium on Wearable Computers (ISWC)","author":"Zhou Yexu","year":"2024","unstructured":"Yexu Zhou, Tobias King, Haibin Zhao, Yiran Huang, Till Riedel, and Michael Beigl. 2024. MLP-HAR: Boosting performance and efficiency of HAR models on edge devices with purely fully connected layers. In 2024 ACM International Symposium on Wearable Computers (ISWC), 133\u2013139."}],"container-title":["ACM Transactions on Computing for Healthcare"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3793553","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T14:26:18Z","timestamp":1775485578000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3793553"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,6]]},"references-count":45,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,4,30]]}},"alternative-id":["10.1145\/3793553"],"URL":"https:\/\/doi.org\/10.1145\/3793553","relation":{},"ISSN":["2691-1957","2637-8051"],"issn-type":[{"value":"2691-1957","type":"print"},{"value":"2637-8051","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,6]]},"assertion":[{"value":"2025-06-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-25","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-06","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}