Open Access Open Access  Restricted Access Subscription Access

A Literature Survey on Implementation of On-Device AI Algorithms for Fall Detection Using Embedded Motion Sensor Data

Ahmed Ali Khan, Ayush Kumar, Mukund Singh, Varun Jain, Dr. Divyashree M

Abstract


Falls among elderly individuals remain one of the major causes of accidental injury and hospitalization worldwide. Recent advancements in Embedded Artificial Intelligence (AI), TinyML, wearable sensing, and edge computing have enabled intelligent fall detection systems capable of operating directly on low-power embedded processors. This paper presents a detailed literature survey of AI-driven embedded fall detection systems focusing on motion sensors, sound analysis, multimodal fusion, TinyML deployment, model optimisation, and commu- nication alert mechanisms. The paper additionally analyses the complete Edge AI workflow involving sensor data acquisition, preprocessing, AI model training, model optimisation, embedded deployment, real-time inference, and emergency communication triggering.

Full Text:

PDF

References


P. Mercier, M. Audette, and C. Paolini, “Ultra low-power, wear- able, accelerated shallow-learning fall detection for elderly at- risk persons,” Smart Health, vol. 33, p. 100498, Sep. 2024, doi: 10.1016/j.smhl.2024.100498.

A. Yasmin, T. Mahmud, S. T. Haque, S. Alamgeer, and A. H. H. Ngu, “Enhancing real-world fall detection using commodity devices: A systematic study,” Sensors, vol. 25, no. 17, p. 5249, Aug. 2025, doi: 10.3390/s25175249.

Y. Li, “Threshold-based smartwatch fall detection for elderly users,” IEEE Access, 2023.

M. S. Al-Rakhami et al., “FallDeF5: A fall detection framework using 5G-based deep gated recurrent unit networks,” IEEE Access, vol. 9, pp. 94299–94308, 2021, doi: 10.1109/ACCESS.2021.3462896.

J. Pan et al., “Human falling detection algorithm based on multisensor data fusion with SVM,” Mobile Information Systems, vol. 2020, p. 8826088, 2020, doi: 10.1155/2020/8826088.

M. J. Al Nahian et al., “Towards an accelerometer-based elderly fall detection system using cross-disciplinary time series features,”

IEEE Access, vol. 9, pp. 39413–39431, 2021, doi: 10.1109/AC-CESS.2021.3056441.

N. Noor and I. K. Park, “A lightweight skeleton-based 3D-CNN for real-time fall detection and action recognition,” in Proc. ICCVW, 2023, pp. 2179–2188, doi: 10.1109/ICCVW60793.2023.00232.

Y. Sanjalawe et al., “Hybrid deep learning for human fall detection: A synergistic approach using YOLOv8 and time-space transformers,” IEEE Access, vol. 13, pp. 41703–41718, 2025, doi: 10.1109/AC-CESS.2025.3542760.

J.Zhangetal.,“Aneffectivedeeplearningframeworkforfalldetection: Model development and study design,” Journal of Medical Internet Research, vol. 26, p. e56750, 2024, doi: 10.2196/56750.

B.BarshanandM.S.Turan,“Anovelheuristicfall-detectionalgorithm based on double thresholding, fuzzy logic, and wearable motion sensor data,” IEEE Internet of Things Journal, vol. 10, no. 20, pp. 17797– 17812, 2023, doi: 10.1109/JIOT.2023.3280060.


Refbacks

  • There are currently no refbacks.