Songzhen Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Songzhen Yang has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Yuan He
Tsinghua University
2 shared publications Yunhao Liu
School of Software, Tsinghua University, Beijing, China
2 shared publicationsLi Yan
Cainiao Network, Alibaba Group, Hangzhou, China
1 shared publicationMeng Jin
School of Software and BNRist, Tsinghua University, Beijing, China
1 shared publication- Meng Jin
School of Artificial Intelligence, Shanghai Jiao Tong University, Shanghai, China
1 shared publication - Xiaolong Zheng
Beijing University of Posts and Telecommunications, Beijing, China
1 shared publication
Latest Publications
- 10 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 22 (10) , 6223-6236
Versatile RFID-Based Sensing: Model, Algorithm, and Applications
The signal phase is one of the most important metrics in RFID-based sensing, which is a useful technique enabling many significant applications. However, existing approaches of RFID-based sensing are often restricted in terms of the sensing capability or accuracy, due to the phase entanglement problem: the phase of the RFID signal is jointly affected by multiple factors, and the change in the signal phase cannot be directly attributed to any one of them. In order to tackle this problem, we propose RF-Prism, a versatile sensing approach that can simultaneously infer multiple physical factors (i.e., location, orientation, and material of targets), purely based on the phase readings. RF-Prism includes a comprehensive model to describe how different physical factors affect the phase of the received signal, and a complete design to disentangle the phase in the multi-frequency and multi-antenna scenario. We implement RF-Prism and evaluate its performance with extensive experiments. The results show that RF-Prism simultaneously achieves a mean localization error of$\text{7.61}~cm$ , a mean orientation error of 9.83 degrees, a mean tracking error of$\text{6.12}~cm$ , and 87.9% material identification accuracy, which outperforms state-of-the-art approaches.- 10 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 22 (10) , 6223-6236
Versatile RFID-Based Sensing: Model, Algorithm, and Applications
The signal phase is one of the most important metrics in RFID-based sensing, which is a useful technique enabling many significant applications. However, existing approaches of RFID-based sensing are often restricted in terms of the sensing capability or accuracy, due to the phase entanglement problem: the phase of the RFID signal is jointly affected by multiple factors, and the change in the signal phase cannot be directly attributed to any one of them. In order to tackle this problem, we propose RF-Prism, a versatile sensing approach that can simultaneously infer multiple physical factors (i.e., location, orientation, and material of targets), purely based on the phase readings. RF-Prism includes a comprehensive model to describe how different physical factors affect the phase of the received signal, and a complete design to disentangle the phase in the multi-frequency and multi-antenna scenario. We implement RF-Prism and evaluate its performance with extensive experiments. The results show that RF-Prism simultaneously achieves a mean localization error of$\text{7.61}~cm$ , a mean orientation error of 9.83 degrees, a mean tracking error of$\text{6.12}~cm$ , and 87.9% material identification accuracy, which outperforms state-of-the-art approaches.- 27 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 20 (4) , 1590-1601
RED: RFID-Based Eccentricity Detection for High-Speed Rotating Machinery
Eccentricity detection is a crucial issue for high-speed rotating machinery, which concerns the stability and safety of the machinery. Conventional techniques in industry for eccentricity detection are mainly based on measuring certain physical indicators, which are costly and hard to deploy. In this paper, we propose RED, a non-intrusive, low-cost, and real-time RFID-based eccentricity detection approach. Differing from the existing RFID-based sensing approaches, RED utilizes the temporal and phase distributions of tag readings as effective features for eccentricity detection. RED includes a Markov chain based model called RUM, which only needs a few sample readings from the tag to make a highly accurate and precise judgement. The design of RED further addresses practical issues, such as parameterizing the RUM model, making it robust to dynamic and noisy environments, and considering how the doppler shift may affect our system. We implement RED with COTS RFID reader and tags, and evaluate its performance across various scenarios. The overall accuracy is 93.6 percent and the detection latency is 0.68 seconds in average.
- 27 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 20 (4) , 1590-1601
RED: RFID-Based Eccentricity Detection for High-Speed Rotating Machinery
Eccentricity detection is a crucial issue for high-speed rotating machinery, which concerns the stability and safety of the machinery. Conventional techniques in industry for eccentricity detection are mainly based on measuring certain physical indicators, which are costly and hard to deploy. In this paper, we propose RED, a non-intrusive, low-cost, and real-time RFID-based eccentricity detection approach. Differing from the existing RFID-based sensing approaches, RED utilizes the temporal and phase distributions of tag readings as effective features for eccentricity detection. RED includes a Markov chain based model called RUM, which only needs a few sample readings from the tag to make a highly accurate and precise judgement. The design of RED further addresses practical issues, such as parameterizing the RUM model, making it robust to dynamic and noisy environments, and considering how the doppler shift may affect our system. We implement RED with COTS RFID reader and tags, and evaluate its performance across various scenarios. The overall accuracy is 93.6 percent and the detection latency is 0.68 seconds in average.