Body Area Networks: Smart IoT and Big Data for Intelligent Health Management. 14th EAI International Conference, BODYNETS 2019, Florence, Italy, October 2-3, 2019, Proceedings

Research Article

A Portable Continuous Wave Radar System to Detect Elderly Fall

  • @INPROCEEDINGS{10.1007/978-3-030-34833-5_1,
        author={Muhammad Ali and Malikeh Ebrahim and Mehmet Yuce},
        title={A Portable Continuous Wave Radar System to Detect Elderly Fall},
        proceedings={Body Area Networks:  Smart IoT and Big Data for Intelligent Health Management. 14th EAI International Conference, BODYNETS 2019, Florence, Italy, October 2-3, 2019, Proceedings},
        proceedings_a={BODYNETS},
        year={2019},
        month={11},
        keywords={Fall detection CWR STFT},
        doi={10.1007/978-3-030-34833-5_1}
    }
    
  • Muhammad Ali
    Malikeh Ebrahim
    Mehmet Yuce
    Year: 2019
    A Portable Continuous Wave Radar System to Detect Elderly Fall
    BODYNETS
    Springer
    DOI: 10.1007/978-3-030-34833-5_1
Muhammad Ali1, Malikeh Ebrahim1, Mehmet Yuce1,*
  • 1: Monash University
*Contact email: mehmet.yuce@monash.edu

Abstract

Fall is the leading cause of death among elderly people worldwide. In this work a low power portable continuous wave radar (CWR) system is proposed to detect elderly fall. This paper presents experimental evaluation of the system to detect human fall motion among various sitting, standing and walking activities. Signals from three subjects with different heights and weights engaged with the different movement activities including walking, sitting, standing and fall in front of the proposed radar system are analyzed. Overall, 60 fall and 180 non-fall activities were recorded. The Short-time Fourier Transform (STFT) is employed to obtain time-frequency Doppler signatures of different human activities. Radar data is analysed by using MATLAB and an algorithm is employed to classify the fall on the basis of analysed data. The results show that the proposed portable CWR can be used to detect fall from non-fall activities with almost 100% accuracy.