Qibin Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Qibin Sun has a total of 126 co-authors. Here are the top co-authors by number of shared publications:
- Kaiping Xue
University of Science and Technology of China, Hefei, Anhui, China
34 shared publications - Jun Lu
School of Cyber Science and Technology, University of Science and Technology of China, Hefei, China
31 shared publications - Yan Chen
University of Science and Technology of China
26 shared publications - Dongheng Zhang
University of Science and Technology of China
25 shared publications - Jian Li
University of Science and Technology of China
20 shared publications - Yang Hu
University of Science and Technology of China
17 shared publications
Latest Publications
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (12) , 23068-23084
Cardiophone: Ubiquitous Cardiovascular Diseases Screening With Smartphone
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (12) , 23068-23084
Cardiophone: Ubiquitous Cardiovascular Diseases Screening With Smartphone
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (11) , 20132-20149
Rethinking Wi-Fi Localization From the Perspective of Wi-Fi Sensing: A Disentangled Representation Learning Approach
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (11) , 20132-20149
Rethinking Wi-Fi Localization From the Perspective of Wi-Fi Sensing: A Disentangled Representation Learning Approach
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (17) , 41043-41057
Toward Practical Learning-Based Indoor Localization in the Real-World Wi-Fi ISAC System
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (17) , 41043-41057
Toward Practical Learning-Based Indoor Localization in the Real-World Wi-Fi ISAC System
- Open Access
- 15 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10  (2) , 1-34
RF-HeartSSL: Self-Supervised Learning for RF-Based Cardiac Sensing
Radio-frequency (RF) sensing has emerged as a promising technique for cardiac monitoring, enabling fully contactless and operation-free measurements. Although recent machine learning approaches have achieved remarkable improvements over traditional signal processing methods, their reliance on supervised training is constrained by the scarcity of annotated RF data, as RF signals are inherently difficult to interpret and hard to annotate manually. This data-scale bottleneck limits the scalability and generalization of existing methods, motivating the need for self-supervised learning (SSL) that can exploit large volumes of unlabeled RF data. However, existing SSL frameworks cannot be directly applied to RF-based cardiac sensing, as the high interference nature of RF signals and the lack of fine-grained cardiac dynamics modeling together can lead to false self-supervision. In this paper, we propose RF-HeartSSL, a self-supervised learning framework that leverages unlabeled RF data to pre-train radio representations for efficient learning in downstream cardiac monitoring tasks. RF-HeartSSL exploits the inherent consistency within radio signals to formulate self-supervised objectives. At the signal level, it models waveform variations driven by signal interference, enforcing representation consistency through contrastive learning. At the physiological level, it constructs a self-temporal alignment strategy that enforces consistent temporal feature extraction along the cardiac progression, enabling the model to capture fine-grained cardiac dynamics. Together, these two objectives enable the model to learn high-fidelity cardiac representations directly from unlabeled RF data. We implement RF-HeartSSL with 3,147 hours of unlabeled RF data for pretraining and evaluate its effectiveness on a large-scale cohort of 7,338 outpatients. The results demonstrate substantial performance improvements in downstream cardiac monitoring tasks, including heart rhythm monitoring and arrhythmia diagnosis, compared with state-of-the-art RF-based cardiac sensing. Moreover, cross-domain evaluations across diverse environments and external clinical validations further confirm the superior robustness and generalization capability of the proposed framework.
- 15 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10  (2) , 1-34
Open AccessRF-HeartSSL: Self-Supervised Learning for RF-Based Cardiac Sensing
Radio-frequency (RF) sensing has emerged as a promising technique for cardiac monitoring, enabling fully contactless and operation-free measurements. Although recent machine learning approaches have achieved remarkable improvements over traditional signal processing methods, their reliance on supervised training is constrained by the scarcity of annotated RF data, as RF signals are inherently difficult to interpret and hard to annotate manually. This data-scale bottleneck limits the scalability and generalization of existing methods, motivating the need for self-supervised learning (SSL) that can exploit large volumes of unlabeled RF data. However, existing SSL frameworks cannot be directly applied to RF-based cardiac sensing, as the high interference nature of RF signals and the lack of fine-grained cardiac dynamics modeling together can lead to false self-supervision. In this paper, we propose RF-HeartSSL, a self-supervised learning framework that leverages unlabeled RF data to pre-train radio representations for efficient learning in downstream cardiac monitoring tasks. RF-HeartSSL exploits the inherent consistency within radio signals to formulate self-supervised objectives. At the signal level, it models waveform variations driven by signal interference, enforcing representation consistency through contrastive learning. At the physiological level, it constructs a self-temporal alignment strategy that enforces consistent temporal feature extraction along the cardiac progression, enabling the model to capture fine-grained cardiac dynamics. Together, these two objectives enable the model to learn high-fidelity cardiac representations directly from unlabeled RF data. We implement RF-HeartSSL with 3,147 hours of unlabeled RF data for pretraining and evaluate its effectiveness on a large-scale cohort of 7,338 outpatients. The results demonstrate substantial performance improvements in downstream cardiac monitoring tasks, including heart rhythm monitoring and arrhythmia diagnosis, compared with state-of-the-art RF-based cardiac sensing. Moreover, cross-domain evaluations across diverse environments and external clinical validations further confirm the superior robustness and generalization capability of the proposed framework.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (8) , 13324-13339
Contactless Arrhythmia Detection via Diversity-Invariant Contrastive mmWave Sensing
Arrhythmias are prevalent cardiac disorders affecting millions worldwide. By analyzing cardiac motion modulated in mmWave reflections, mmWave sensing is emerging as a promising contactless revolution in arrhythmia detection compared to conventional contact-based methods. However, the fundamental bottleneck of existing mmWave sensing methods is their restriction to controlled laboratory settings with small-scale cohorts, limiting generalization to real-world populations. This limitation arises because mmWave signals undergo complex signal transformations during propagation, resulting in an explosion of signal diversity across large populations in real-world scenarios. Such diversity significantly complicates the direct recognition of arrhythmia. In this paper, we theoretically analyze the mechanism and impact of mmWave cardiac signal diversity. Leveraging the inherent transformation properties of mmWave signals, we propose a Diversity-Invariant Contrastive mmWave Sensing framework, which learns invariant features robust to complex signal transformations encountered in real-world scenarios. We evaluate our method in a practical, clinically-oriented scenario involving a large-scale population of 7,338 subjects, achieving an average F1-score of 0.8241 across four common arrhythmias. These results demonstrate that our method effectively bridges the diversity gap, representing a significant step toward practical clinical deployment of contactless arrhythmia detection via mmWave sensing.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (8) , 13324-13339
Contactless Arrhythmia Detection via Diversity-Invariant Contrastive mmWave Sensing
Arrhythmias are prevalent cardiac disorders affecting millions worldwide. By analyzing cardiac motion modulated in mmWave reflections, mmWave sensing is emerging as a promising contactless revolution in arrhythmia detection compared to conventional contact-based methods. However, the fundamental bottleneck of existing mmWave sensing methods is their restriction to controlled laboratory settings with small-scale cohorts, limiting generalization to real-world populations. This limitation arises because mmWave signals undergo complex signal transformations during propagation, resulting in an explosion of signal diversity across large populations in real-world scenarios. Such diversity significantly complicates the direct recognition of arrhythmia. In this paper, we theoretically analyze the mechanism and impact of mmWave cardiac signal diversity. Leveraging the inherent transformation properties of mmWave signals, we propose a Diversity-Invariant Contrastive mmWave Sensing framework, which learns invariant features robust to complex signal transformations encountered in real-world scenarios. We evaluate our method in a practical, clinically-oriented scenario involving a large-scale population of 7,338 subjects, achieving an average F1-score of 0.8241 across four common arrhythmias. These results demonstrate that our method effectively bridges the diversity gap, representing a significant step toward practical clinical deployment of contactless arrhythmia detection via mmWave sensing.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Radar Systems
- Vol. 4, 1349-1360
mmFallBbox: An Effective and Efficient Fall Detection Framework With a Comprehensive Benchmark
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Radar Systems
- Vol. 4, 1349-1360
mmFallBbox: An Effective and Efficient Fall Detection Framework With a Comprehensive Benchmark
- Open Access
- 2 December 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 9  (4) , 1-30
Robust HRV Monitoring via Massive Radio Sensing
Heart Rate Variability (HRV) is a crucial biomarker in health monitoring and disease management. Radio sensing has emerged as a promising contactless alternative, addressing the limitations of conventional contact-based techniques. However, a major challenge of existing approaches is their poor generalization performance in real-world deployments. This arises from the inherent sensitivity of radio signals to environmental variations, causing intrinsic shifts in signal distribution and representation. As a result, current methods struggle to adapt to different deployment conditions, leading to performance degradation in real-world applications where complex environments are unavoidable. In this paper, we systematically analyze the generalization challenge posed by environmental variations from the perspective of statistical signal modeling and formulate it as an estimation problem under a global environmental distribution. Inspired by the law of large numbers, we assume that assembling a sufficiently large number of environment-variant samples allows the empirical risk to approximate the true risk, thereby yielding a robust estimator. Accordingly, we propose a novel Massive Radio Sensing framework that leverages massive environment-variant signal sampling and a structured deep learning optimization strategy to statistically derive a robust HRV estimator. We evaluate our method across a broad spectrum of real-world deployment scenarios. Specifically, testing on 30 participants across 32 application-oriented environments—including home and workplace settings—demonstrates a 29.7% improvement in performance compared to the current state-of-the-art method. To further assess clinical generalizability, we evaluate our method on 130 inpatients across 8 distinct hospital environments, achieving a 41.7% performance improvement. These results highlight the effectiveness of our approach and its strong potential for real-world deployment in radio-based HRV monitoring.
- 2 December 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 9  (4) , 1-30
Open AccessRobust HRV Monitoring via Massive Radio Sensing
Heart Rate Variability (HRV) is a crucial biomarker in health monitoring and disease management. Radio sensing has emerged as a promising contactless alternative, addressing the limitations of conventional contact-based techniques. However, a major challenge of existing approaches is their poor generalization performance in real-world deployments. This arises from the inherent sensitivity of radio signals to environmental variations, causing intrinsic shifts in signal distribution and representation. As a result, current methods struggle to adapt to different deployment conditions, leading to performance degradation in real-world applications where complex environments are unavoidable. In this paper, we systematically analyze the generalization challenge posed by environmental variations from the perspective of statistical signal modeling and formulate it as an estimation problem under a global environmental distribution. Inspired by the law of large numbers, we assume that assembling a sufficiently large number of environment-variant samples allows the empirical risk to approximate the true risk, thereby yielding a robust estimator. Accordingly, we propose a novel Massive Radio Sensing framework that leverages massive environment-variant signal sampling and a structured deep learning optimization strategy to statistically derive a robust HRV estimator. We evaluate our method across a broad spectrum of real-world deployment scenarios. Specifically, testing on 30 participants across 32 application-oriented environments—including home and workplace settings—demonstrates a 29.7% improvement in performance compared to the current state-of-the-art method. To further assess clinical generalizability, we evaluate our method on 130 inpatients across 8 distinct hospital environments, achieving a 41.7% performance improvement. These results highlight the effectiveness of our approach and its strong potential for real-world deployment in radio-based HRV monitoring.
- Open Access
- 2 December 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 9  (4) , 1-33
Finding Order in Chaos: Learning Disentangled Features for mmWave Cardiac Sensing
Recent advances have demonstrated the great potential of millimeter-wave (mmWave) signals for contactless cardiac sensing, enabling various applications such as heart rate variability (HRV) tracking and arrhythmia detection. However, the absence of structural characterization in mmWave cardiac signals, combined with complex interference, makes it difficult to separate cardiac rhythm, cardiac pattern, and external interference. As a result, existing methods predominantly process these entangled components jointly. This missing step of feature disentanglement, which is fundamental to biosignal analysis, severely limits performance and hinders the translation of mmWave cardiac sensing into clinical practice. In this work, we propose the first disentangled feature learning framework for contactless cardiac mmWave sensing. The key innovation lies in the design of a variational feature disentanglement model, which embeds structural priors of mmWave cardiac signals to construct an inductive bias that facilitates effective disentanglement. This design enables the disentanglement of three key components from signals: cardiac-irrelevant interference features, intrinsic cardiac rhythm features, and intrinsic cardiac pattern features, thereby supporting effective cardiac sensing. We evaluate our framework on a large-scale clinical dataset comprising 7,090 outpatients. Experimental results demonstrate superior performance compared to baseline methods, achieving 27.96% and 28.42% average improvements in HRV tracking and arrhythmia detection tasks, respectively, thus highlighting its strong potential to bridge the gap between mmWave sensing technology and real-world clinical applications.
- 2 December 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 9  (4) , 1-33
Open AccessFinding Order in Chaos: Learning Disentangled Features for mmWave Cardiac Sensing
Recent advances have demonstrated the great potential of millimeter-wave (mmWave) signals for contactless cardiac sensing, enabling various applications such as heart rate variability (HRV) tracking and arrhythmia detection. However, the absence of structural characterization in mmWave cardiac signals, combined with complex interference, makes it difficult to separate cardiac rhythm, cardiac pattern, and external interference. As a result, existing methods predominantly process these entangled components jointly. This missing step of feature disentanglement, which is fundamental to biosignal analysis, severely limits performance and hinders the translation of mmWave cardiac sensing into clinical practice. In this work, we propose the first disentangled feature learning framework for contactless cardiac mmWave sensing. The key innovation lies in the design of a variational feature disentanglement model, which embeds structural priors of mmWave cardiac signals to construct an inductive bias that facilitates effective disentanglement. This design enables the disentanglement of three key components from signals: cardiac-irrelevant interference features, intrinsic cardiac rhythm features, and intrinsic cardiac pattern features, thereby supporting effective cardiac sensing. We evaluate our framework on a large-scale clinical dataset comprising 7,090 outpatients. Experimental results demonstrate superior performance compared to baseline methods, achieving 27.96% and 28.42% average improvements in HRV tracking and arrhythmia detection tasks, respectively, thus highlighting its strong potential to bridge the gap between mmWave sensing technology and real-world clinical applications.
- 21 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23  (2) , 3316-3330
FESCAT: Function Secret Sharing Based Efficient Secure Collaborative Analysis of Time Series Data
Time series data analysis, employing dynamic time warping (DTW) algorithms, has a wide range of applications in fields such as medicine and economics. Given the widespread distribution of data across different domains, integrating and analyzing these datasets through outsourced cloud computing can enhance analytics, though privacy concerns arise. Privacy preserving data analysis, underpinned by secure multi-party computing, emerges as a crucial approach to address this challenge. However, existing efforts face high communication costs and increased interactions, resulting in significant efficiency constraints in practical applications. In this paper, we propose a function secret sharing (FSS)-based framework for secure collaborative analysis of time series data using the DTW algorithm. Utilizing the distributed comparison function, we develop efficient building blocks with minimal online interaction and communication, enhancing the practicability of security protocols. To address the challenges of FSS key generation due to uncertain computational topology when cascading multiple distances, we adopt a modular design and decompose the analysis process into several critical modules. Furthermore, our framework efficiently supports various constraint methods for DTW. We implement and evaluate our framework using publicly available datasets. The results demonstrate a significant reduction in communication costs and the number of interactions during the online phase.
- 21 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23  (2) , 3316-3330
FESCAT: Function Secret Sharing Based Efficient Secure Collaborative Analysis of Time Series Data
Time series data analysis, employing dynamic time warping (DTW) algorithms, has a wide range of applications in fields such as medicine and economics. Given the widespread distribution of data across different domains, integrating and analyzing these datasets through outsourced cloud computing can enhance analytics, though privacy concerns arise. Privacy preserving data analysis, underpinned by secure multi-party computing, emerges as a crucial approach to address this challenge. However, existing efforts face high communication costs and increased interactions, resulting in significant efficiency constraints in practical applications. In this paper, we propose a function secret sharing (FSS)-based framework for secure collaborative analysis of time series data using the DTW algorithm. Utilizing the distributed comparison function, we develop efficient building blocks with minimal online interaction and communication, enhancing the practicability of security protocols. To address the challenges of FSS key generation due to uncertain computational topology when cascading multiple distances, we adopt a modular design and decompose the analysis process into several critical modules. Furthermore, our framework efficiently supports various constraint methods for DTW. We implement and evaluate our framework using publicly available datasets. The results demonstrate a significant reduction in communication costs and the number of interactions during the online phase.