TY - GEN
T1 - LoDiHAR
T2 - 21st Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2024
AU - Sun, Xiaoqi
AU - Wang, Yanwen
AU - Zhang, Chenwei
AU - Wang, Zheng
AU - Shi, Xiaokang
AU - Zheng, Yuanqing
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Human Activity Recognition has been extensively applied to fulfill tasks such as fall detection, human-computer interaction, virtual reality, etc. Existing radio frequency-based HAR methods, although overcoming limitations of wearable-, visual-, and acoustic-based sensing technology, still suffer from high costs and low efficiency, which limits their pervasive use. In this paper, we propose LoDiHAR, a low-cost, distributed HAR system leveraging Radio Frequency Identification technology. LoDiHAR employs low-cost and fully programmable commercial wireless components, providing full access to the PHY samples of the backscattered signals, in which signal phases can be extracted to infer different activities. Different from COTS RFID systems that adopt a polling interrogation scheme, LoDiHAR supports a distributed sensing scheme, which profiles human activities more efficiently. LoDiHAR addresses a series of technical challenges such as accurate phase extraction from backscattered signals, asynchronous distributed RF data fusion and insufficient training data. A Conditional Generative Adversarial Network framework combined with a Transformer model is designed for accurate time-series activity classification. LoDiHAR demonstrates proficiency in recognizing eight types of human activities across diverse environments, achieving an accuracy of up to 94.9% while only costing 10% of the mainstream COTS RFID systems.
AB - Human Activity Recognition has been extensively applied to fulfill tasks such as fall detection, human-computer interaction, virtual reality, etc. Existing radio frequency-based HAR methods, although overcoming limitations of wearable-, visual-, and acoustic-based sensing technology, still suffer from high costs and low efficiency, which limits their pervasive use. In this paper, we propose LoDiHAR, a low-cost, distributed HAR system leveraging Radio Frequency Identification technology. LoDiHAR employs low-cost and fully programmable commercial wireless components, providing full access to the PHY samples of the backscattered signals, in which signal phases can be extracted to infer different activities. Different from COTS RFID systems that adopt a polling interrogation scheme, LoDiHAR supports a distributed sensing scheme, which profiles human activities more efficiently. LoDiHAR addresses a series of technical challenges such as accurate phase extraction from backscattered signals, asynchronous distributed RF data fusion and insufficient training data. A Conditional Generative Adversarial Network framework combined with a Transformer model is designed for accurate time-series activity classification. LoDiHAR demonstrates proficiency in recognizing eight types of human activities across diverse environments, achieving an accuracy of up to 94.9% while only costing 10% of the mainstream COTS RFID systems.
KW - data augmentation
KW - distributed sensing
KW - low-cost
KW - RFID
UR - http://www.scopus.com/inward/record.url?scp=105002314556&partnerID=8YFLogxK
U2 - 10.1109/SECON64284.2024.10934903
DO - 10.1109/SECON64284.2024.10934903
M3 - Conference article published in proceeding or book
AN - SCOPUS:105002314556
T3 - Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops
BT - 2024 21st Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2024
PB - IEEE Computer Society
Y2 - 2 December 2024 through 4 December 2024
ER -