TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition
RFID-based human activity recognition (HAR) attracts attention due to its convenience, non-invasiveness, and privacy protection. Existing RFID-based HAR methods use modeling, CNN, or LSTM to extract features effectively. Still, they have shortcomings: 1) requiring complex hand-crafted data cleaning...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2024-02-01
|
| Series: | Defence Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214914723000508 |
| _version_ | 1827325523379879936 |
|---|---|
| author | Yi Liu Weiqing Huang Shang Jiang Bobai Zhao Shuai Wang Siye Wang Yanfang Zhang |
| author_facet | Yi Liu Weiqing Huang Shang Jiang Bobai Zhao Shuai Wang Siye Wang Yanfang Zhang |
| author_sort | Yi Liu |
| collection | DOAJ |
| description | RFID-based human activity recognition (HAR) attracts attention due to its convenience, non-invasiveness, and privacy protection. Existing RFID-based HAR methods use modeling, CNN, or LSTM to extract features effectively. Still, they have shortcomings: 1) requiring complex hand-crafted data cleaning processes and 2) only addressing single-person activity recognition based on specific RF signals. To solve these problems, this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM. This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing. Concretely, we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes single-human activities and human-to-human interactions. Compared with existing CNN- and LSTM-based methods, the Transformer-based method has more data fitting power, generalization, and scalability. Furthermore, using RF signals, our method achieves an excellent classification effect on human behavior-based classification tasks. Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy (99.1%). The dataset we collected for detecting RFID-based indoor human activities will be published. |
| first_indexed | 2024-03-07T14:29:29Z |
| format | Article |
| id | doaj.art-ceebf1857d584300b7e69a8408c9948e |
| institution | Directory Open Access Journal |
| issn | 2214-9147 |
| language | English |
| last_indexed | 2024-03-07T14:29:29Z |
| publishDate | 2024-02-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Defence Technology |
| spelling | doaj.art-ceebf1857d584300b7e69a8408c9948e2024-03-06T05:26:52ZengKeAi Communications Co., Ltd.Defence Technology2214-91472024-02-0132619628TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognitionYi Liu0Weiqing Huang1Shang Jiang2Bobai Zhao3Shuai Wang4Siye Wang5Yanfang Zhang6Institute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China; Corresponding author.Institute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, ChinaInstitute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, ChinaSchool of Information Management, Beijing Information Science and Technology University, Beijing, ChinaInstitute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, ChinaInstitute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China; Corresponding author.Institute of Information Engineering Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences, Beijing, ChinaRFID-based human activity recognition (HAR) attracts attention due to its convenience, non-invasiveness, and privacy protection. Existing RFID-based HAR methods use modeling, CNN, or LSTM to extract features effectively. Still, they have shortcomings: 1) requiring complex hand-crafted data cleaning processes and 2) only addressing single-person activity recognition based on specific RF signals. To solve these problems, this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM. This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing. Concretely, we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes single-human activities and human-to-human interactions. Compared with existing CNN- and LSTM-based methods, the Transformer-based method has more data fitting power, generalization, and scalability. Furthermore, using RF signals, our method achieves an excellent classification effect on human behavior-based classification tasks. Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy (99.1%). The dataset we collected for detecting RFID-based indoor human activities will be published.http://www.sciencedirect.com/science/article/pii/S2214914723000508Human activity recognitionRFIDTransformer |
| spellingShingle | Yi Liu Weiqing Huang Shang Jiang Bobai Zhao Shuai Wang Siye Wang Yanfang Zhang TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition Defence Technology Human activity recognition RFID Transformer |
| title | TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition |
| title_full | TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition |
| title_fullStr | TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition |
| title_full_unstemmed | TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition |
| title_short | TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition |
| title_sort | transtm a device free method based on time streaming multiscale transformer for human activity recognition |
| topic | Human activity recognition RFID Transformer |
| url | http://www.sciencedirect.com/science/article/pii/S2214914723000508 |
| work_keys_str_mv | AT yiliu transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT weiqinghuang transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT shangjiang transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT bobaizhao transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT shuaiwang transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT siyewang transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition AT yanfangzhang transtmadevicefreemethodbasedontimestreamingmultiscaletransformerforhumanactivityrecognition |