Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks
Accidents caused by falls among the elderly have become a significant social issue, making fall detection systems increasingly needed. Fall detection systems such as internet of things (IoT) devices must be affordable and compact because they must be installed in various locations around the house,...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/2/546 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589243510161408 |
|---|---|
| author | Hyeongwon Cho Soongyu Kang Yunseong Sim Seongjoo Lee Yunho Jung |
| author_facet | Hyeongwon Cho Soongyu Kang Yunseong Sim Seongjoo Lee Yunho Jung |
| author_sort | Hyeongwon Cho |
| collection | DOAJ |
| description | Accidents caused by falls among the elderly have become a significant social issue, making fall detection systems increasingly needed. Fall detection systems such as internet of things (IoT) devices must be affordable and compact because they must be installed in various locations around the house, such as bedrooms, living rooms, and bathrooms. In this study, we propose a lightweight fall detection method using a continuous-wave (CW) radar sensor and a binarized neural network (BNN) to meet these requirements. We used a CW radar sensor, which is more affordable than other types of radar sensors, and employed a BNN with binarized features and parameters to reduce memory usage and make the system lighter. The proposed method distinguishes movements using micro-Doppler signatures, and spectrogram is binarized as an input to the BNN. The proposed method achieved 93.1% accuracy in binary classification of five fall actions and six non-fall actions. The memory requirements for storing parameters were reduced to 11.9 KB, representing a reduction of up to 99.9% compared with previous studies. |
| format | Article |
| id | doaj-art-26a0388e5d3f4d319437cc33ed06ef6e |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-26a0388e5d3f4d319437cc33ed06ef6e2025-01-24T13:19:46ZengMDPI AGApplied Sciences2076-34172025-01-0115254610.3390/app15020546Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural NetworksHyeongwon Cho0Soongyu Kang1Yunseong Sim2Seongjoo Lee3Yunho Jung4School of Electronics and Information Engineering, Korea Aerospace University, Goyang 10540, Republic of KoreaDepartment of Smart Air Mobility, Korea Aerospace University, Goyang 10540, Republic of KoreaDepartment of Smart Air Mobility, Korea Aerospace University, Goyang 10540, Republic of KoreaDepartment of Electrical Engineering, Sejong University, Seoul 05006, Republic of KoreaSchool of Electronics and Information Engineering, Korea Aerospace University, Goyang 10540, Republic of KoreaAccidents caused by falls among the elderly have become a significant social issue, making fall detection systems increasingly needed. Fall detection systems such as internet of things (IoT) devices must be affordable and compact because they must be installed in various locations around the house, such as bedrooms, living rooms, and bathrooms. In this study, we propose a lightweight fall detection method using a continuous-wave (CW) radar sensor and a binarized neural network (BNN) to meet these requirements. We used a CW radar sensor, which is more affordable than other types of radar sensors, and employed a BNN with binarized features and parameters to reduce memory usage and make the system lighter. The proposed method distinguishes movements using micro-Doppler signatures, and spectrogram is binarized as an input to the BNN. The proposed method achieved 93.1% accuracy in binary classification of five fall actions and six non-fall actions. The memory requirements for storing parameters were reduced to 11.9 KB, representing a reduction of up to 99.9% compared with previous studies.https://www.mdpi.com/2076-3417/15/2/546fall detectioninternet of thingslightweightcontinuous wave radarshort-time Fourier transformmicro-Doppler signature |
| spellingShingle | Hyeongwon Cho Soongyu Kang Yunseong Sim Seongjoo Lee Yunho Jung Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks Applied Sciences fall detection internet of things lightweight continuous wave radar short-time Fourier transform micro-Doppler signature |
| title | Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks |
| title_full | Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks |
| title_fullStr | Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks |
| title_full_unstemmed | Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks |
| title_short | Fall Detection Based on Continuous Wave Radar Sensor Using Binarized Neural Networks |
| title_sort | fall detection based on continuous wave radar sensor using binarized neural networks |
| topic | fall detection internet of things lightweight continuous wave radar short-time Fourier transform micro-Doppler signature |
| url | https://www.mdpi.com/2076-3417/15/2/546 |
| work_keys_str_mv | AT hyeongwoncho falldetectionbasedoncontinuouswaveradarsensorusingbinarizedneuralnetworks AT soongyukang falldetectionbasedoncontinuouswaveradarsensorusingbinarizedneuralnetworks AT yunseongsim falldetectionbasedoncontinuouswaveradarsensorusingbinarizedneuralnetworks AT seongjoolee falldetectionbasedoncontinuouswaveradarsensorusingbinarizedneuralnetworks AT yunhojung falldetectionbasedoncontinuouswaveradarsensorusingbinarizedneuralnetworks |