Energy-Efficient TinyML Approach for Wearable Fall Detection on Edge Devices Using Spatial-Temporal Deep Learning

Fall detection systems are critical for elderly care; however, cross-dataset generalization and practical edge deployment are challenges in existing approaches. This paper presents an efficient wearable fall detection system based on CNN-LSTM that achieves robust performance across multiple benchmark datasets, with real-time inference on resource-constrained microcontroller units (MCU) while maintaining low energy consumption. The proposed architecture combines convolutional layers with long short-term memory cells to capture spatial-temporal patterns in tri-axial accelerometer signals and their RMS magnitude. Using LOSO validation on the KFall dataset, the model achieves a subject-averaged accuracy of 99.3% and an F1-score of 98.97%. For cross-dataset validation, zero-shot transfer to SisFall achieved 98.3% accuracy and 97.9% F1-score without retraining, representing approximately 1.0% performance degradation under controlled laboratory conditions. The trained model is successfully deployed on an ESP32-S3 MCU, achieving an inference latency of 113.6 ms per 4.0-s window with an average power consumption of 5.78 mA, enabling up to 7 days of continuous operation or approximately 3 weeks under typical daily usage cycles. The proposed system offers a highly practical, energy-efficient baseline that paves the way for future real-world elderly monitoring applications by successfully integrating efficient MCU execution with robust cross-dataset transfer on simulated falls.

CNN-LSTM; Fall detection; Real-time inference; TinyML; Wearable sensors

Abdou, A., Mascret, Q., Gurve, D., Gosselin, B., & Krishnan, S. (2025). Embedded tinyml approach for fall detection in geriatric care. https://doi.org/10.36227/techrxiv.174918043.39615584/v1

Afuan, L., & Isnanto, R. (2025). A comparative study of machine learning algorithms for fall detection in technology-based healthcare system: Analyzing svm, knn, decision tree, random forest, lstm, and cnn. E3S Web of Conferences, 605, 03051. https://doi.org/10.1051/e3sconf/202560503051

Amir, N., Dziyauddin, R., Mohamed, N., Ismail, N., Kaidi, H., Ahmad, N., & Izhar, M. (2024). Fall detection system using wearable sensor devices and machine learning: A review. https://doi.org/10.36227/techrxiv.171084921.16728034/v1

Arif, M., & Rashid, M. (2025). A literature review on model conversion, inference, and learning strategies in edgeml with tinyml deployment. Computers, Materials & Continua, 83(1), 13–64. https://doi.org/10.32604/cmc.2025.062819

Aziz, O., Musngi, M., Park, E., Mori, G., & Robinovitch, S. (2017). A comparison of accuracy of fall detection algorithms (threshold-based vs. machine learning) using waist-mounted tri-axial accelerometer signals from a comprehensive set of falls and non-fall trials. Medical & Biological Engineering & Computing, 55(1), 45–55. https://doi.org/10.1007/s11517-016-1504-y

Benoit, A., Escriba, C., Gauchard, D., Esteve, A., & Rossi, C. (2024). Analyzing and comparing deep learning models on an arm 32 bits microcontroller for pre-impact fall detection. IEEE Sensors Journal, 24(7), 11829–11842. https://doi.org/10.1109/JSEN.2024.3364249

David, R., Duke, J., Jain, A., Reddi, V., Jeffries, N., Li, J., Kreeger, N., Nappier, I., Natraj, M., Regev, S., Rhodes, R., Wang, T., & Warden, P. (2021). Tensorflow lite micro: Embedded machine learning on tinyml systems. http://arxiv.org/abs/2010.08678

Fula, V., & Moreno, P. (2024). Wrist-based fall detection: Towards generalization across datasets. Sensors, 24(5), 1679. https://doi.org/10.3390/s24051679

Gorce, P., & Jacquier-Bret, J. (2025). Fall detection in elderly people: A systematic review of ambient assisted living and smart home-related technology performance. Sensors, 25(21), 6540. https://doi.org/10.3390/s25216540

Guo, P., & Nakayama, M. (2025). A feature engineering method for smartphone-based fall detection. Sensors, 25(20), 6500. https://doi.org/10.3390/s25206500

Haslam-Larmer, L., Donnelly, C., Auais, M., Woo, K., & DePaul, V. (2021). Early mobility after fragility hip fracture: A mixed methods embedded case study. BMC Geriatrics, 21(1), 181. https://doi.org/10.1186/s12877-021-02083-3

Hu, J., Cheng, F., Liu, M., Xu, X., & Li, X. (2025a). Microfallnet: A lightweight model for real-time fall detection on smart wristbands. Pervasive and Mobile Computing, 102046. https://doi.org/10.1016/j.pmcj.2025.102046

Hu, X., Yu, S., Zheng, J., Fang, Z., Zhao, Z., & Qu, X. (2025b). A hybrid cnn-lstm model for involuntary fall detection using wrist-worn sensors. Advanced Engineering Informatics, 65, 103178. https://doi.org/10.1016/j.aei.2025.103178

Hussain, F., Hussain, F., Ehatisham-ul-Haq, M., & Azam, M. (2019). Activity-aware fall detection and recognition based on wearable sensors. IEEE Sensors Journal, 19(12), 4528–4536. https://doi.org/10.1109/JSEN.2019.2898891

Jiang, Z., Al-Qaness, M., Al-Alimi, D., Ewees, A., Abd Elaziz, M., Dahou, A., & Helmi, A. (2024). Fall detection systems for internet of medical things based on wearable sensors: A review. IEEE Internet of Things Journal, 11(21), 34797–34810. https://doi.org/10.1109/JIOT.2024.3421336

Kakara, R., Bergen, G., Burns, E., & Stevens, M. (2023). Nonfatal and fatal falls among adults aged 65 years — united states, 2020–2021. MMWR. Morbidity and Mortality Weekly Report, 72(35), 938–943. https://doi.org/10.15585/mmwr.mm7235a1

Kallimani, R., Pai, K., Raghuwanshi, P., Iyer, S., & L?opez, O. (2023). Tinyml: Tools, applications, challenges, and future research directions. Multimedia Tools and Applications, 83(10), 29015–29045. https://doi.org/10.1007/s11042-023-16740-9

Kausar, F., Mesbah, M., Iqbal, W., Ahmad, A., & Sayyed, I. (2024). Fall detection in the elderly using different machine learning algorithms with optimal window size. Mobile Networks and Applications, 29(2), 413–423. https://doi.org/10.1007/s11036-023-02215-6

Koo, B., Yu, X., Lee, S., Yang, S., Kim, D., Xiong, S., & Kim, Y. (2023). Tinyfallnet: A lightweight pre-impact fall detection model. Sensors, 23(20), 8459. https://doi.org/10.3390/s23208459

Lau, X., Connie, T., Goh, M., & Lau, S. (2022). Fall detection and motion analysis using visual approaches. International Journal of Technology, 13(6), 1173. https://doi.org/10.14716/ijtech.v13i6.5840

Li, H., Shrestha, A., Heidari, H., Le Kernec, J., & Fioranelli, F. (2020). Bi-lstm network for multimodal continuous human activity recognition and fall detection. IEEE Sensors Journal, 20(3), 1191–1201. https://doi.org/10.1109/JSEN.2019.2946095

Liu, J., Li, X., Huang, S., Chao, R., Cao, Z., Wang, S., Wang, A., & Liu, L. (2023). A review of wearable sensors based fall-related recognition systems. Engineering Applications of Artificial Intelligence, 121, 105993. https://doi.org/10.1016/j.engappai.2023.105993

Mao, W.-L., Wang, C.-C., Chou, P.-H., Liu, K.-C., & Tsao, Y. (2024). Meckd: Deep learning-based fall detection in multilayer mobile edge computing with knowledge distillation. IEEE Sensors Journal, 24(24), 42195–42209. https://doi.org/10.1109/JSEN.2024.3456577

Maruf, M., Haque, M., Hasan, M., Farhan, M., & Islam, A. (2025). State-of-the-art review on fall prediction among older adults: Exploring edge devices as a promising approach for the future. Measurement: Sensors, 39, 101878. https://doi.org/10.1016/j.measen.2025.101878

Mohan, D., Al-Hamid, D., Chong, P., Sudheera, K., Gutierrez, J., Chan, H., & Li, H. (2024). Artificial intelligence and iot in elderly fall prevention: A review. IEEE Sensors Journal, 24(4), 4181–4198. https://doi.org/10.1109/JSEN.2023.3344605

Naeim, M., Chung, G., Lee, I., Tiang, J., & Tan, S. (2023). A mobile iot-based elderly monitoring system for senior safety. International Journal of Technology, 14(6), 1185. https://doi.org/10.14716/ijtech.v14i6.6634

Naja, S., Din Makhlouf, M., & Chehab, M. (2017). An ageing world of the 21st century: A literature review. International Journal Of Community Medicine And Public Health, 4(12), 4363. https://doi.org/10.18203/2394-6040.ijcmph20175306

Nizam, Y., & Jamil, M. (2020). Classification of daily life activities for human fall detection: A systematic review of the techniques and approaches. In Challenges and trends in multimodal fall detection for healthcare (pp. 137–179). Springer International Publishing. https://doi.org/10.1007/978-3-030-38748-8_7

Noury, N., Fleury, A., Rumeau, P., Bourke, A., Laighin, G., Rialle, V., & Lundy, J. (2007). Fall detection - principles and methods. 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1663–1666. https://doi.org/10.1109/IEMBS.2007.4352627

Pereira, C., De Oliveira, E., & De Souza, A. (2024). Machine learning applied to edge computing and wearable devices for healthcare: Systematic mapping of the literature. Sensors, 24(19), 6322. https://doi.org/10.3390/s24196322

Poh, S.-C., Tan, Y.-F., Cheong, S.-N., Ooi, C.-P., & Tan, W.-H. (2019). Anomaly detection for home activity based on sequence pattern. International Journal of Technology, 10(7), 1276. https://doi.org/10.14716/ijtech.v10i7.3230

Rubenstein, L. (2006). Falls in older people: Epidemiology, risk factors and strategies for prevention. Age and Ageing, 35(suppl 2), ii37–ii41. https://doi.org/10.1093/ageing/afl084

Sahni, S., Jain, S., & Saritha, S. (2025). A novel ensemble model for fall detection: Leveraging cnn and bilstm with channel and temporal attention. Automatika, 66(2), 103–116. https://doi.org/10.1080/00051144.2025.2450553

Sehairi, K., & Chouireb, F. (2023). Implementation of motion detection methods on embedded systems: A performance comparison. International Journal of Technology, 14(3), 510. https://doi.org/10.14716/ijtech.v14i3.5950

Silva, C., Casilari, E., & Garc??a-Berm?udez, R. (2024). Cross-dataset evaluation of wearable fall detection systems using data from real falls and long-term monitoring of daily life. Measurement, 235, 114992. https://doi.org/10.1016/j.measurement.2024.114992

Sucerquia, A., L?opez, J., & Vargas-Bonilla, J. (2017). Sisfall: A fall and movement dataset. Sensors, 17(1), 198. https://doi.org/10.3390/s17010198

Verma, S., Willetts, J., Corns, H., Marucci-Wellman, H., Lombardi, D., & Courtney, T. (2016). Falls and fall-related injuries among community-dwelling adults in the united states. PLOS ONE, 11(3), e0150939. https://doi.org/10.1371/journal.pone.0150939

Wang, X., Ellul, J., & Azzopardi, G. (2020). Elderly fall detection systems: A literature survey. Frontiers in Robotics and AI, 7, 71. https://doi.org/10.3389/frobt.2020.00071

Xu, T., Se, H., & Liu, J. (2021). A fusion fall detection algorithm combining threshold-based method and convolutional neural network. Microprocessors and Microsystems, 82, 103828. https://doi.org/10.1016/j.micpro.2021.103828

Xu, T., Zhou, Y., & Zhu, J. (2018). New advances and challenges of fall detection systems: A survey. Applied Sciences, 8(3), 418. https://doi.org/10.3390/app8030418

Yahyati, C., Lamaakal, I., Maleh, Y., Makkaoui, K., Ouahbi, I., Almousa, M., & Abd El-Latif, A. (2025). A systematic review of state-of-the-art tinyml applications in healthcare, education, and transportation. IEEE Access, 1–1. https://doi.org/10.1109/ACCESS.2025.3633575

Yu, X., Jang, J., & Xiong, S. (2021). A large-scale open motion dataset (kfall) and benchmark algorithms for detecting pre-impact fall of the elderly using wearable inertial sensors. Frontiers in Aging Neuroscience, 13, 692865. https://doi.org/10.3389/fnagi.2021.692865

Yu, X., Park, S., Kim, D., Kim, E., Kim, J., Kim, W., An, Y., & Xiong, S. (2023). A practical wearable fall detection system based on tiny convolutional neural networks. Biomedical Signal Processing and Control, 86, 105325. https://doi.org/10.1016/j.bspc.2023.105325

Zhou, H., Zhang, X., Feng, Y., Zhang, T., & Xiong, L. (2025). Efficient human activity recognition on edge devices using deepconv lstm architectures. Scientific Reports, 15(1), 13830. https://doi.org/10.1038/s41598-025-98571-2