dc.contributor.author | Mohtadifar, Masoud | |
dc.contributor.author | Cheffena, Michael | |
dc.contributor.author | Pourafzal, Alireza | |
dc.date.accessioned | 2023-03-08T08:26:50Z | |
dc.date.available | 2023-03-08T08:26:50Z | |
dc.date.created | 2022-04-22T13:06:10Z | |
dc.date.issued | 2022 | |
dc.identifier.citation | Sensors. 2022, 22 (9), . | en_US |
dc.identifier.issn | 1424-8220 | |
dc.identifier.uri | https://hdl.handle.net/11250/3056916 | |
dc.description.abstract | In this work, a hybrid radio frequency (RF)- and acoustic-based activity recognition system was developed to demonstrate the advantage of combining two non-invasive sensors in Human Activity Recognition (HAR) systems and smart assisted living. We used a hybrid approach, employing RF and acoustic signals to recognize falling, walking, sitting on a chair, and standing up from a chair. To our knowledge, this is the first work that attempts to use a mixture of RF and passive acoustic signals for Human Activity Recognition purposes. We conducted experiments in the lab environment using a Vector Network Analyzer measuring the 2.4 GHz frequency band and a microphone array. After recording data, we extracted the Mel-spectrogram feature of the audio data and the Doppler shift feature of the RF measurements. We fed these features to six classification algorithms. Our result shows that using a hybrid acoustic- and radio-based method increases the accuracy of recognition compared to just using only one kind of sensory data and shows the possibility of expanding for a variety of other different activities that can be recognized. We demonstrate that by using a hybrid method, the recognition accuracy increases in all classification algorithms. Among these classifiers, five of them achieve over 98% recognition accuracy. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | MDPI | en_US |
dc.rights | Navngivelse 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by/4.0/deed.no | * |
dc.title | Acoustic-and Radio-Frequency-Based Human Activity Recognition | en_US |
dc.title.alternative | Acoustic-and Radio-Frequency-Based Human Activity Recognition | en_US |
dc.type | Peer reviewed | en_US |
dc.type | Journal article | en_US |
dc.description.version | publishedVersion | en_US |
dc.source.pagenumber | 21 | en_US |
dc.source.volume | 22 | en_US |
dc.source.journal | Sensors | en_US |
dc.source.issue | 9 | en_US |
dc.identifier.doi | 10.3390/s22093125 | |
dc.identifier.cristin | 2018414 | |
dc.relation.project | Norges forskningsråd: 300638 | en_US |
cristin.ispublished | true | |
cristin.fulltext | original | |
cristin.qualitycode | 1 | |