Mengyuan Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mengyuan Wang has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Hongbo Jiang
Hunan University
5 shared publications - Ping Zhao
Donghua University
2 shared publications - Schahram Dustdar
TU Wien
2 shared publications - Zhu Xiao
College of Computer Science and Electronic Engineering, Hunan University, Changsha, China
2 shared publications - Daibo Liu
College of Computer Science and Electronic Engineering, Hunan University, Changsha, China
1 shared publication - Geyong Min
University of Exeter
1 shared publication
Latest Publications
- 18 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (5) , 5953-5968
$\mathsf {RobustHealth}$RobustHealth: Non-Interactive Privacy-Preserving System for Heterogeneous Mobile Health Diagnosis
The mobile health (mHealth) system, leveraging mobile edge computing, can monitor health status and provide diagnosis. However, due to the privacy of medical data and the resource limitations of mobile devices, patients are unable to access diagnostic services provided by untrusted servers in real-time. Existing schemes present significant challenges in private heterogeneous data aggregation, model training and inference in the presence of malicious participants, and expensive resource consumption. To address these issues, in this paper, we propose a non-interactive privacy-preserving system with the naive Bayesian model, i.e.,$\mathsf {RobustHealth}$ , for heterogeneous mHealth diagnosis. Specifically, we extract homogeneous features from heterogeneous datasets to enable efficient encrypted aggregation. We propose a novel private model training algorithm with enhanced security to against collusion-then-differential attacks. We develop a novel non-interactive private model inference algorithm using minimal lightweight cryptographic primitives, designed for patients under unstable network environments. We provide formal security proofs for our system using the Universal Composable (UC) framework. To validate the performance of$\mathsf {RobustHealth}$ , we conduct extensive experiments on real-world heterogeneous datasets, and compared with related works. The results demonstrate a$\mathbf {4.37\%}$ improvement in model accuracy, along with significant reductions in computational and communication overheads of$\mathbf {21.18\times }$ and$\mathbf {4.24\times }$ , respectively.- 18 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (5) , 5953-5968
$\mathsf {RobustHealth}$RobustHealth: Non-Interactive Privacy-Preserving System for Heterogeneous Mobile Health Diagnosis
The mobile health (mHealth) system, leveraging mobile edge computing, can monitor health status and provide diagnosis. However, due to the privacy of medical data and the resource limitations of mobile devices, patients are unable to access diagnostic services provided by untrusted servers in real-time. Existing schemes present significant challenges in private heterogeneous data aggregation, model training and inference in the presence of malicious participants, and expensive resource consumption. To address these issues, in this paper, we propose a non-interactive privacy-preserving system with the naive Bayesian model, i.e.,$\mathsf {RobustHealth}$ , for heterogeneous mHealth diagnosis. Specifically, we extract homogeneous features from heterogeneous datasets to enable efficient encrypted aggregation. We propose a novel private model training algorithm with enhanced security to against collusion-then-differential attacks. We develop a novel non-interactive private model inference algorithm using minimal lightweight cryptographic primitives, designed for patients under unstable network environments. We provide formal security proofs for our system using the Universal Composable (UC) framework. To validate the performance of$\mathsf {RobustHealth}$ , we conduct extensive experiments on real-world heterogeneous datasets, and compared with related works. The results demonstrate a$\mathbf {4.37\%}$ improvement in model accuracy, along with significant reductions in computational and communication overheads of$\mathbf {21.18\times }$ and$\mathbf {4.24\times }$ , respectively.- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (16) , 32889-32903
Dynamic Personalized Federated Learning via Representation-Driven Clustering
Clustering Federated Learning (CFL) promotes knowledge sharing by adaptively grouping similar clients while solving the Non-IID problem to provide personalized models with high generalization. However, there are several problems needed to be considered before the practical deployment of CFL: 1) Practical clients often have limited communication capabilities. 2) Dynamically evolving data distributions of clients may leads to unreasonable clustering. 3) Asynchronization will cause clients belonging to the same cluster to fall apart. The above three problems will reduce the convergence speed and damage the accuracy of the model. In this work, we propose a Dynamic Representation-driven Clustering Federated Learning framework (DReCFL) to solve the above three problems. Specifically, DReCFL replaces complex model parameters with data representation for communicating to reduce the communication pressure. In order to gain a reasonable clustering, DReCFL utilizes a Dynamic Client Fuzzy Clustering (DCFC) algorithm, we proposed, to adapt to evolving data distributions. Finally, DReCFL leverages an Adaptive Clustering Threshold (ACT) mechanism, we designed, to craft clustering thresholds based on training progress, ensuring the reliability of client clustering results under asynchrony. Extensive experimental results demonstrate that DReCFL efficiently adapts to the addition of new clients and changes in data distribution, while reducing communication overhead by 22.70%-83.33% compared to state-of-the-art (SOTA) personalized federated learning methods.
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (16) , 32889-32903
Dynamic Personalized Federated Learning via Representation-Driven Clustering
Clustering Federated Learning (CFL) promotes knowledge sharing by adaptively grouping similar clients while solving the Non-IID problem to provide personalized models with high generalization. However, there are several problems needed to be considered before the practical deployment of CFL: 1) Practical clients often have limited communication capabilities. 2) Dynamically evolving data distributions of clients may leads to unreasonable clustering. 3) Asynchronization will cause clients belonging to the same cluster to fall apart. The above three problems will reduce the convergence speed and damage the accuracy of the model. In this work, we propose a Dynamic Representation-driven Clustering Federated Learning framework (DReCFL) to solve the above three problems. Specifically, DReCFL replaces complex model parameters with data representation for communicating to reduce the communication pressure. In order to gain a reasonable clustering, DReCFL utilizes a Dynamic Client Fuzzy Clustering (DCFC) algorithm, we proposed, to adapt to evolving data distributions. Finally, DReCFL leverages an Adaptive Clustering Threshold (ACT) mechanism, we designed, to craft clustering thresholds based on training progress, ensuring the reliability of client clustering results under asynchrony. Extensive experimental results demonstrate that DReCFL efficiently adapts to the addition of new clients and changes in data distribution, while reducing communication overhead by 22.70%-83.33% compared to state-of-the-art (SOTA) personalized federated learning methods.
- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (2) , 662-676
Echoes of Fingertip: Unveiling POS Terminal Passwords Through Wi-Fi Beamforming Feedback
Recent years, point-of-sale (POS) terminals are no longer limited to wired connections, with many relying on Wi-Fi for data transmission. Although Wi-Fi offers the convenience of wireless connectivity, it introduces significant security vulnerabilities. This work presents a non-intrusive method for eavesdropping POS passwords via Wi-Fi sensing, named${\mathsf {BeamThief}}$ . Instead of conventional Wi-Fi Channel State Information (CSI) readings, our approach employs Wi-Fi Beamforming Feedback Information (BFI) for an eavesdropping attack. Compared to CSI, which can only be extracted through intruding into the Access Point (AP) or from a limited selection of commercial Wi-Fi cards (e.g., Intel-5300), BFI readings can be more readily obtained from a broad array of commercial Wi-Fi devices. A key technological contribution of${\mathsf {BeamThief}}$ is the development of an analysis model for predicting finger motion trajectories. This model is based on the physical relationship between BFI readings and finger motion, thus eliminating the need for extensive labeled training data. Furthermore, we employ Maximum Ratio Combining (MRC) to enhance the BFI series, ensuring performance across various scenarios. We implement${\mathsf {BeamThief}}$ using everyday commercial Wi-Fi devices and conduct a series of experiments to assess the impact of this attack. Experimental results demonstrate that${\mathsf {BeamThief}}$ achieves an accuracy rate 79$\%$ in inferring 6-digit POS passwords within the top-100 attempts.- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (2) , 662-676
Echoes of Fingertip: Unveiling POS Terminal Passwords Through Wi-Fi Beamforming Feedback
Recent years, point-of-sale (POS) terminals are no longer limited to wired connections, with many relying on Wi-Fi for data transmission. Although Wi-Fi offers the convenience of wireless connectivity, it introduces significant security vulnerabilities. This work presents a non-intrusive method for eavesdropping POS passwords via Wi-Fi sensing, named${\mathsf {BeamThief}}$ . Instead of conventional Wi-Fi Channel State Information (CSI) readings, our approach employs Wi-Fi Beamforming Feedback Information (BFI) for an eavesdropping attack. Compared to CSI, which can only be extracted through intruding into the Access Point (AP) or from a limited selection of commercial Wi-Fi cards (e.g., Intel-5300), BFI readings can be more readily obtained from a broad array of commercial Wi-Fi devices. A key technological contribution of${\mathsf {BeamThief}}$ is the development of an analysis model for predicting finger motion trajectories. This model is based on the physical relationship between BFI readings and finger motion, thus eliminating the need for extensive labeled training data. Furthermore, we employ Maximum Ratio Combining (MRC) to enhance the BFI series, ensuring performance across various scenarios. We implement${\mathsf {BeamThief}}$ using everyday commercial Wi-Fi devices and conduct a series of experiments to assess the impact of this attack. Experimental results demonstrate that${\mathsf {BeamThief}}$ achieves an accuracy rate 79$\%$ in inferring 6-digit POS passwords within the top-100 attempts.- 29 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (5) , 6408-6422
Accurately Estimating Frequencies of Relations With Relation Privacy Preserving in Decentralized Networks
Abundant valuable knowledge can be obtained by learning frequencies of relations in a decentralized network, which benefits various further complex tasks, such as range query and commodity recommendation. Nonetheless, counting frequencies in original networks can reveal sensitive data and pose a risk to individual privacy, specifically relation privacy. Current privacy notions do not fully preserve both the privacy of relations' values and existence. In this paper, we introduce an enhanced privacy notion, relation local differential privacy (relation LDP), which provides comprehensive preservation in relation privacy. However, a significant amount of noise in perturbed networks often leads to severe errors in the accuracy of relation frequencies. To obtain accurate frequencies, we propose a framework called frequency estimation based on combination (Fest-C), which decomposes relation frequencies into two independent parts, the total frequency of relations and relative proportions. Binning relations into hyper-relations, Fest-C reduces errors of total frequency and relative proportions, respectively, and then estimates frequencies by logically multiplying them. Finally, we rigorously prove that Fest-C satisfies relation LDP. Our experiments on various datasets confirm the high accuracy of Fest-C in estimating relation frequencies with a much lower time overhead compared to competitors.
- 29 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (5) , 6408-6422
Accurately Estimating Frequencies of Relations With Relation Privacy Preserving in Decentralized Networks
Abundant valuable knowledge can be obtained by learning frequencies of relations in a decentralized network, which benefits various further complex tasks, such as range query and commodity recommendation. Nonetheless, counting frequencies in original networks can reveal sensitive data and pose a risk to individual privacy, specifically relation privacy. Current privacy notions do not fully preserve both the privacy of relations' values and existence. In this paper, we introduce an enhanced privacy notion, relation local differential privacy (relation LDP), which provides comprehensive preservation in relation privacy. However, a significant amount of noise in perturbed networks often leads to severe errors in the accuracy of relation frequencies. To obtain accurate frequencies, we propose a framework called frequency estimation based on combination (Fest-C), which decomposes relation frequencies into two independent parts, the total frequency of relations and relative proportions. Binning relations into hyper-relations, Fest-C reduces errors of total frequency and relative proportions, respectively, and then estimates frequencies by logically multiplying them. Finally, we rigorously prove that Fest-C satisfies relation LDP. Our experiments on various datasets confirm the high accuracy of Fest-C in estimating relation frequencies with a much lower time overhead compared to competitors.
- 10 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (3) , 2351-2366
RoPriv: Road Network-Aware Privacy-Preserving Framework in Spatial Crowdsourcing
Spatial Crowdsourcing (SC) has been an indispensable Location-based Service where the SC server assigns tasks to workers based on the locations of task requesters and workers, raising strong privacy concerns. Limited by the computational and time complexity, existing works prefer differential privacy-based methods to protect location privacy. However, most differential privacy-based works ignore the road network, perturbing locations on two-dimensional plane, resulting in more failures in tasks and moreover extensive privacy disclosure in practice. This paper aims to implement a multi-task assignment with both high utility and efficiency while protecting the location privacy of both task requesters and workers on road networks. Specifically, we design a Road Network-aware Exponential Mechanism and propose an Obfuscated Locations Selection algorithm to guarantee location privacy of all participants and extensive privacy. Then, we propose region distance. Based on this, we further formulate multi-task assignment as a Binary Linear Programming problem and a utility-aware optimization problem. We solve the first problem to obtain optimal efficiency and then propose a utility-aware optimization algorithm for the second problem to improve the utility. Our experiments demonstrate sufficient and stable privacy guarantee and the well-performance on both utility and efficiency of our framework.
- 10 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (3) , 2351-2366
RoPriv: Road Network-Aware Privacy-Preserving Framework in Spatial Crowdsourcing
Spatial Crowdsourcing (SC) has been an indispensable Location-based Service where the SC server assigns tasks to workers based on the locations of task requesters and workers, raising strong privacy concerns. Limited by the computational and time complexity, existing works prefer differential privacy-based methods to protect location privacy. However, most differential privacy-based works ignore the road network, perturbing locations on two-dimensional plane, resulting in more failures in tasks and moreover extensive privacy disclosure in practice. This paper aims to implement a multi-task assignment with both high utility and efficiency while protecting the location privacy of both task requesters and workers on road networks. Specifically, we design a Road Network-aware Exponential Mechanism and propose an Obfuscated Locations Selection algorithm to guarantee location privacy of all participants and extensive privacy. Then, we propose region distance. Based on this, we further formulate multi-task assignment as a Binary Linear Programming problem and a utility-aware optimization problem. We solve the first problem to obtain optimal efficiency and then propose a utility-aware optimization algorithm for the second problem to improve the utility. Our experiments demonstrate sufficient and stable privacy guarantee and the well-performance on both utility and efficiency of our framework.
- 4 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 29 (5) , 2228-2241
A Utility-Aware General Framework With Quantifiable Privacy Preservation for Destination Prediction in LBSs
Destination prediction plays an important role as the basis for a variety of location-based services (LBSs). However, it poses many threats to users’ location privacy. Most related work ignores privacy preservation in destination prediction. Few studies focus on specific kinds of privacy-preserving destination prediction algorithms and thus are not applicable to other prediction methods. Furthermore, the third party involved in these studies is a potential privacy threat. Additionally, another line of related work regarding LBSs neither guarantees the utility of the predicted results nor provides quantifiable privacy preservation. To this end, in this paper, we propose a general framework that can provide quantifiable privacy preservation and obtain a trade-off between the privacy and the utility of the predicted results by utilizing differential privacy and a neural network model. Specifically, it first adopts a specially designed differential privacy to construct a data-driven privacy-preserving model that formulates the relationship between injected noise and privacy preservation. Then, it combines a Recurrent Neural Network and Multi-hill Climbing to add fine-grained noise to obtain the trade-off between the privacy preservation and the utility of the predicted results. Our extensive experiments on real-world datasets validate that the proposed framework can be applied to different prediction methods, provide quantifiable location privacy preservation, and guarantee the utility of the predicted results simultaneously.
- 4 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 29 (5) , 2228-2241
A Utility-Aware General Framework With Quantifiable Privacy Preservation for Destination Prediction in LBSs
Destination prediction plays an important role as the basis for a variety of location-based services (LBSs). However, it poses many threats to users’ location privacy. Most related work ignores privacy preservation in destination prediction. Few studies focus on specific kinds of privacy-preserving destination prediction algorithms and thus are not applicable to other prediction methods. Furthermore, the third party involved in these studies is a potential privacy threat. Additionally, another line of related work regarding LBSs neither guarantees the utility of the predicted results nor provides quantifiable privacy preservation. To this end, in this paper, we propose a general framework that can provide quantifiable privacy preservation and obtain a trade-off between the privacy and the utility of the predicted results by utilizing differential privacy and a neural network model. Specifically, it first adopts a specially designed differential privacy to construct a data-driven privacy-preserving model that formulates the relationship between injected noise and privacy preservation. Then, it combines a Recurrent Neural Network and Multi-hill Climbing to add fine-grained noise to obtain the trade-off between the privacy preservation and the utility of the predicted results. Our extensive experiments on real-world datasets validate that the proposed framework can be applied to different prediction methods, provide quantifiable location privacy preservation, and guarantee the utility of the predicted results simultaneously.