Weikuan Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Weikuan Yu has a total of 55 co-authors. Here are the top co-authors by number of shared publications:
- Bofan Li
Department of Computer Science, Florida State University, Tallahassee, Florida, USA
2 shared publications - Jeffrey S. Vetter
Oak Ridge National Laboratory, Oak Ridge, TN, USA
2 shared publications - Xin Liu
Department of Computer Science, Florida State University, Tallahassee, Florida, USA
2 shared publications Zhuo Liu
Auburn Univ, Dept Comp Sci & Software Engn, Auburn, AL 36849 USA
2 shared publicationsAlberto Miranda
Barcelona Supercomputing Center
1 shared publicationAmit Kumar Nath
Florida State Univ, Dept Comp Sci, Tallahassee, FL 32306 USA
1 shared publication
Latest Publications
- Open Access
- 30 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (3) , 1-26
BFMScan: Enabling Explicit Angle-Resolved Sensing via Beamforming Feedback Matrix
Commodity Wi-Fi sensing has gained significant traction by leveraging the Beamforming Feedback Matrix (BFM) available on standard devices. However, existing BFM-based approaches suffer from a fundamental limitation: they primarily rely on relative temporal variations contributed by the overall environment (e.g., amplitude/phase variations of BFM ratios) instead of explicitly extracting physically grounded parameters. Consequently, these methods are intrinsically motion-dependent, rendering them blind to static targets, and spatially under-constrained, necessitating strong initialization priors (e.g., known starting positions), multi-device coordination, or data-driven learning to resolve spatial ambiguity. In this paper, we present BFMScan, the first sensing framework to achieve explicit angle-resolved sensing using only the downlink BFM from a single commodity Wi-Fi device pair. Rather than relying on indirect temporal variations for spatial inference, we mathematically derive the subspace structure from compressed BFM. This theoretical foundation enables a subspace-based algorithm that reconstructs high-fidelity Angle-of-Departure (AoD) profiles from BFM, achieving absolute spatial observability. This capability allows BFMScan to detect both static and moving targets and resolve their spatial directions without requiring external initial position. Furthermore, we extend this angle-domain separability to the spatiotemporal domain to robustly distinguish multiple concurrent targets motions. We implement BFMScan on a single pair of commercial off-the-shelf (COTS) Wi-Fi devices and evaluate it across diverse scenarios, including single-user device-based angular tracking, two-user human interaction, and three-user respiration monitoring. Experimental results demonstrate that BFMScan enables accurate angle-resolved sensing from BFM, supports multiple targets, and achieves robust performance across complex real-world environments.
- 30 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (3) , 1-26
Open AccessBFMScan: Enabling Explicit Angle-Resolved Sensing via Beamforming Feedback Matrix
Commodity Wi-Fi sensing has gained significant traction by leveraging the Beamforming Feedback Matrix (BFM) available on standard devices. However, existing BFM-based approaches suffer from a fundamental limitation: they primarily rely on relative temporal variations contributed by the overall environment (e.g., amplitude/phase variations of BFM ratios) instead of explicitly extracting physically grounded parameters. Consequently, these methods are intrinsically motion-dependent, rendering them blind to static targets, and spatially under-constrained, necessitating strong initialization priors (e.g., known starting positions), multi-device coordination, or data-driven learning to resolve spatial ambiguity. In this paper, we present BFMScan, the first sensing framework to achieve explicit angle-resolved sensing using only the downlink BFM from a single commodity Wi-Fi device pair. Rather than relying on indirect temporal variations for spatial inference, we mathematically derive the subspace structure from compressed BFM. This theoretical foundation enables a subspace-based algorithm that reconstructs high-fidelity Angle-of-Departure (AoD) profiles from BFM, achieving absolute spatial observability. This capability allows BFMScan to detect both static and moving targets and resolve their spatial directions without requiring external initial position. Furthermore, we extend this angle-domain separability to the spatiotemporal domain to robustly distinguish multiple concurrent targets motions. We implement BFMScan on a single pair of commercial off-the-shelf (COTS) Wi-Fi devices and evaluate it across diverse scenarios, including single-user device-based angular tracking, two-user human interaction, and three-user respiration monitoring. Experimental results demonstrate that BFMScan enables accurate angle-resolved sensing from BFM, supports multiple targets, and achieves robust performance across complex real-world environments.
- Open Access
- 16 March 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (1) , 1-26
MURAL-Fi: M ulti- U ser R espiration A uthentication L everaging Wi Fi
Integrating WiFi communication with vital signs sensing, particularly under the emerging 802.11bf standard, offers a promising pathway toward next-generation authentication systems, providing non-intrusive, continuous, and scalable security. However, current WiFi-based respiration authentication approaches are limited to single-user scenarios, whereas existing multi-user respiration monitoring technologies either lack identity awareness or require users to remain completely stationary. In this paper, we present MURAL-Fi, the first system to achieve multi-user respiration authentication using WiFi, addressing three critical requirements: multi-user support, identity-aware capability, and involuntary-motion tolerance. Unlike conventional methods that monitor the movement of the highest-energy point within the respiration frequency band of the WiFi spectrum, MURAL-Fi introduces a novel baseline region variation exploration technique. This technique identifies a spatial region of points corresponding to the chest cavities and extracts respiration signals by analyzing strength variations within these regions. Specifically, we iteratively employ baseline region identification coupled with interference suppression utilizing spatial spectral information derived from WiFi signals. This iterative refinement progressively enhances the precision of chest-cavity localization and robustly mitigates various interferences, enabling identity-aware respiration pattern extraction despite involuntary body movements. Subsequently, we perform multi-level feature analysis to derive user-specific respiratory characteristics and utilize a triplet loss-based Siamese Neural Network to authenticate users. Through extensive experiments with 24 participants across diverse environments, we demonstrate that MURAL-Fi achieves high authentication accuracy, maintaining superior performance across challenging scenarios, including resistance to mimic attacks.
- 16 March 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (1) , 1-26
Open AccessMURAL-Fi: M ulti- U ser R espiration A uthentication L everaging Wi Fi
Integrating WiFi communication with vital signs sensing, particularly under the emerging 802.11bf standard, offers a promising pathway toward next-generation authentication systems, providing non-intrusive, continuous, and scalable security. However, current WiFi-based respiration authentication approaches are limited to single-user scenarios, whereas existing multi-user respiration monitoring technologies either lack identity awareness or require users to remain completely stationary. In this paper, we present MURAL-Fi, the first system to achieve multi-user respiration authentication using WiFi, addressing three critical requirements: multi-user support, identity-aware capability, and involuntary-motion tolerance. Unlike conventional methods that monitor the movement of the highest-energy point within the respiration frequency band of the WiFi spectrum, MURAL-Fi introduces a novel baseline region variation exploration technique. This technique identifies a spatial region of points corresponding to the chest cavities and extracts respiration signals by analyzing strength variations within these regions. Specifically, we iteratively employ baseline region identification coupled with interference suppression utilizing spatial spectral information derived from WiFi signals. This iterative refinement progressively enhances the precision of chest-cavity localization and robustly mitigates various interferences, enabling identity-aware respiration pattern extraction despite involuntary body movements. Subsequently, we perform multi-level feature analysis to derive user-specific respiratory characteristics and utilize a triplet loss-based Siamese Neural Network to authenticate users. Through extensive experiments with 24 participants across diverse environments, we demonstrate that MURAL-Fi achieves high authentication accuracy, maintaining superior performance across challenging scenarios, including resistance to mimic attacks.
- 17 January 2026
- Published by Springer Nature
- p. 47-58
Spatial-ViLT: Enhancing Visual Spatial Reasoning Through Multi-task Learning
- 17 January 2026
- Published by Springer Nature
- p. 47-58
Spatial-ViLT: Enhancing Visual Spatial Reasoning Through Multi-task Learning
- Open Access
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 695-707
Spatio-Temporal Resource Control for Cloud-Native GPU Provisioning
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 695-707
Open AccessSpatio-Temporal Resource Control for Cloud-Native GPU Provisioning
- 14 September 2024
- Published by Springer Nature in GeoInformatica
- Vol. 29 (2) , 255-274
LENS: label sparsity-tolerant adversarial learning on spatial deceptive reviews
- 14 September 2024
- Published by Springer Nature in GeoInformatica
- Vol. 29 (2) , 255-274
LENS: label sparsity-tolerant adversarial learning on spatial deceptive reviews
- 10 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 33 (12) , 3929-3941
PhaST: Hierarchical Concurrent Log-Free Skip List for Persistent Memory
Skip list (skiplist) is a competitive index structure that offers superior concurrency and excellent performance but with high memory overhead and low access locality. Emerging persistent memory (PM) technologies present an opportunity to mitigate the capacity constraint of DRAM. However, data consistency on PM typically results in excessive write overhead. In addition, fast concurrent access to an index is critical to the throughput on high-end contemporary computer systems. In this article, we propose a Partitioned HierArchical SkiplisT called PhaST , which can simultaneously reduce the skiplist height and improve its access locality, through its hierarchy of component structures, while enabling fast parallel recovery in case of failure. To ensure high concurrency and fast data consistency, we also have developed writelock-free concurrent insert and log-free atomic split. Furthermore, we have developed a durable lock-free concurrent search that can discern transient structural inconsistencies and deliver highly concurrent read operations. We have conducted an extensive evaluation of PhaST compared to state-of-the-art studies such as NV-Skiplist, wB+-Tree, FPTree, and FAST-FAIR. Our evaluation results show PhaST outperforms other indexing structures by up to 4.05× and 2.87× in single-threaded inserts and searches, and 1.56× and 2.62× in concurrent inserts and searches.
- 10 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 33 (12) , 3929-3941
PhaST: Hierarchical Concurrent Log-Free Skip List for Persistent Memory
Skip list (skiplist) is a competitive index structure that offers superior concurrency and excellent performance but with high memory overhead and low access locality. Emerging persistent memory (PM) technologies present an opportunity to mitigate the capacity constraint of DRAM. However, data consistency on PM typically results in excessive write overhead. In addition, fast concurrent access to an index is critical to the throughput on high-end contemporary computer systems. In this article, we propose a Partitioned HierArchical SkiplisT called PhaST , which can simultaneously reduce the skiplist height and improve its access locality, through its hierarchy of component structures, while enabling fast parallel recovery in case of failure. To ensure high concurrency and fast data consistency, we also have developed writelock-free concurrent insert and log-free atomic split. Furthermore, we have developed a durable lock-free concurrent search that can discern transient structural inconsistencies and deliver highly concurrent read operations. We have conducted an extensive evaluation of PhaST compared to state-of-the-art studies such as NV-Skiplist, wB+-Tree, FPTree, and FAST-FAIR. Our evaluation results show PhaST outperforms other indexing structures by up to 4.05× and 2.87× in single-threaded inserts and searches, and 1.56× and 2.62× in concurrent inserts and searches.
- 17 January 2020
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 35 (1) , 4-26
Ad Hoc File Systems for High-Performance Computing
- 17 January 2020
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 35 (1) , 4-26
Ad Hoc File Systems for High-Performance Computing
- 1 July 2019
- Published by Elsevier in Parallel Computing
- Vol. 85, 66-78
Multivariate modeling and two-level scheduling of analytic queries
- 1 July 2019
- Published by Elsevier in Parallel Computing
- Vol. 85, 66-78
Multivariate modeling and two-level scheduling of analytic queries
- 4 August 2017
- Published by Springer Nature in Cluster Computing
- Vol. 21 (2) , 1275-1287
HALO: a fast and durable disk write cache using phase change memory
- 4 August 2017
- Published by Springer Nature in Cluster Computing
- Vol. 21 (2) , 1275-1287
HALO: a fast and durable disk write cache using phase change memory
- 3 December 2015
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 30 (2) , 241-256
Enhance parallel input/output with cross-bundle aggregation
The exponential growth of computing power on leadership scale computing platforms imposes grand challenge to scientific applications’ input/output (I/O) performance. To bridge the performance gap between computation and I/O, various parallel I/O libraries have been developed and adopted by computer scientists. These libraries enhance the I/O parallelism by allowing multiple processes to concurrently access the shared data set. Meanwhile, they are integrated with a set of I/O optimization strategies such as data sieving and two-phase I/O to better exploit the supplied bandwidth of the underlying parallel file system. Most of these techniques are optimized for the access on a single bundle of variables generated by the scientific applications during the I/O phase, which is stored in the form of file. Few of these techniques focus on cross-bundle I/O optimizations. In this article, we investigate the potential benefit from cross-bundle I/O aggregation. Based on the analysis of the I/O patterns of a mission-critical scientific application named the Goddard Earth Observing System, version 5 (GEOS-5), we propose a Bundle-based PARallel Aggregation (BPAR) framework with three partitioning schemes to improve its I/O performance as well as the I/O performance of a broad range of other scientific applications. Our experiment result reveals that BPAR can deliver 2.1× I/O performance improvement over the baseline GEOS-5, and it is very promising in accelerating scientific applications’ I/O performance on various computing platforms.
- 3 December 2015
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 30 (2) , 241-256
Enhance parallel input/output with cross-bundle aggregation
The exponential growth of computing power on leadership scale computing platforms imposes grand challenge to scientific applications’ input/output (I/O) performance. To bridge the performance gap between computation and I/O, various parallel I/O libraries have been developed and adopted by computer scientists. These libraries enhance the I/O parallelism by allowing multiple processes to concurrently access the shared data set. Meanwhile, they are integrated with a set of I/O optimization strategies such as data sieving and two-phase I/O to better exploit the supplied bandwidth of the underlying parallel file system. Most of these techniques are optimized for the access on a single bundle of variables generated by the scientific applications during the I/O phase, which is stored in the form of file. Few of these techniques focus on cross-bundle I/O optimizations. In this article, we investigate the potential benefit from cross-bundle I/O aggregation. Based on the analysis of the I/O patterns of a mission-critical scientific application named the Goddard Earth Observing System, version 5 (GEOS-5), we propose a Bundle-based PARallel Aggregation (BPAR) framework with three partitioning schemes to improve its I/O performance as well as the I/O performance of a broad range of other scientific applications. Our experiment result reveals that BPAR can deliver 2.1× I/O performance improvement over the baseline GEOS-5, and it is very promising in accelerating scientific applications’ I/O performance on various computing platforms.