Xiaoyan Hu
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
Xiaoyan Hu has a total of 92 co-authors. Here are the top co-authors by number of shared publications:
- Hua Wu
School of Cyber Science and Engineering, Southeast University, Nanjing, China
26 shared publications - Guang Cheng
School of Cyber Science and Engineering, Southeast University, Nanjing 211189, China
24 shared publications - Ruidong Li
Institute of Science and Engineering, Kanazawa University, Kakuma, Kanazawa, Japan
14 shared publications Guang Cheng
6 shared publicationsHua Wu
School of Cyber Science and Engineering, Southeast University, Nanjing, China
3 shared publications- Ke Ding
School of Cyber Science and Engineering, Southeast University, Nanjing, China
3 shared publications
Latest Publications
- 11 November 2026
- Published by Elsevier in Computers & Security
- Vol. 173
vEcho: From passive verification to proactive discovery through evidence-grounded vulnerability auditing
- 11 November 2026
- Published by Elsevier in Computers & Security
- Vol. 173
vEcho: From passive verification to proactive discovery through evidence-grounded vulnerability auditing
- 26 September 2026
- Published by Springer Nature in Applied Intelligence
- Vol. 56 (15) , 453
Visualized anomaly detection on Bitcoin transaction based on the E-GTNet fusion model and graph database
- 26 September 2026
- Published by Springer Nature in Applied Intelligence
- Vol. 56 (15) , 453
Visualized anomaly detection on Bitcoin transaction based on the E-GTNet fusion model and graph database
- 6 August 2026
- Published by Elsevier in Computer Networks
- Vol. 288
AdaDoHTool: A test-time adaptive framework for identifying DoH tunneling tools with unknown resolvers
- 6 August 2026
- Published by Elsevier in Computer Networks
- Vol. 288
AdaDoHTool: A test-time adaptive framework for identifying DoH tunneling tools with unknown resolvers
- 20 May 2026
- Published by Elsevier in Computer Networks
- Vol. 285
FSG-NID: Early network intrusion detection via flow segment graph analysis
- 20 May 2026
- Published by Elsevier in Computer Networks
- Vol. 285
FSG-NID: Early network intrusion detection via flow segment graph analysis
- 18 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 5001-5016
Data-Efficient Cross-Domain Few-Shot Website Fingerprinting With Unsupervised Domain Adaptation
Website fingerprinting (WF) attacks identify Torencrypted websites but struggle with cross-domain scenarios due to traffic distribution shifts. The existing few-shot WF attacks address the cross-domain problem with excessive auxiliary data, significantly reducing deployment efficiency. This work proposes UDA-WF, a data-efficient few-shot WF with Unsupervised Domain Adaptation (UDA). UDA-WF first pre-trains the feature extractor with limited auxiliary data in the source website domain. Then, it extracts the invariant feature space by computing the intersection of the source and target feature spaces through the unsupervised domain adaptation with the softmatch mechanism. Finally, UDA-WF fine-tunes the feature extractor and a single-layer perceptron to extract the discriminative unique feature space of the target website domain. We evaluate UDA-WF on our WF dataset collected over multiple months. UDA-WF significantly overcomes the cross-domain problem while reducing auxiliary data requirements by 95% and pre-training bootstrap time by 99% compared to the State-Of-The-Art (SOTA) methods. UDA-WF achieves an accuracy of 97.37% under the 20-shot setting in the closed-world scenario and outperforms SOTA methods. To further demonstrate the model’s adaptability to diverse real-world requirements, we validate it on the DF and Wang datasets, achieving accuracies exceeding 92% and 94%, respectively. Moreover, the results show that our UDA-WF is more resilient to concept drift and robust to WF defense.
- 18 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 5001-5016
Data-Efficient Cross-Domain Few-Shot Website Fingerprinting With Unsupervised Domain Adaptation
Website fingerprinting (WF) attacks identify Torencrypted websites but struggle with cross-domain scenarios due to traffic distribution shifts. The existing few-shot WF attacks address the cross-domain problem with excessive auxiliary data, significantly reducing deployment efficiency. This work proposes UDA-WF, a data-efficient few-shot WF with Unsupervised Domain Adaptation (UDA). UDA-WF first pre-trains the feature extractor with limited auxiliary data in the source website domain. Then, it extracts the invariant feature space by computing the intersection of the source and target feature spaces through the unsupervised domain adaptation with the softmatch mechanism. Finally, UDA-WF fine-tunes the feature extractor and a single-layer perceptron to extract the discriminative unique feature space of the target website domain. We evaluate UDA-WF on our WF dataset collected over multiple months. UDA-WF significantly overcomes the cross-domain problem while reducing auxiliary data requirements by 95% and pre-training bootstrap time by 99% compared to the State-Of-The-Art (SOTA) methods. UDA-WF achieves an accuracy of 97.37% under the 20-shot setting in the closed-world scenario and outperforms SOTA methods. To further demonstrate the model’s adaptability to diverse real-world requirements, we validate it on the DF and Wang datasets, achieving accuracies exceeding 92% and 94%, respectively. Moreover, the results show that our UDA-WF is more resilient to concept drift and robust to WF defense.
- 18 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (11) , 24823-24837
Toward Evolvable IoT Device-Type Identification Using Few-Shot Incremental Learning
The proliferation of Internet of Things (IoT) devices has profoundly transformed various industries. However, with the continual emergence of new IoT device types, their extensive heterogeneity and weak security have posed significant challenges for network management and security. Despite existing research excelling at classifying IoT traffic, they have yet to consider incremental updates and timely identification, which are critical for early device management and security in IoT networks. As a solution, we present an evolvable attention prototypical network (EAPN), a novel and evolvable IoT device identification model that supports adapting to new IoT devices with limited traffic. The EAPN extracts traffic features from only a few dozen packets and resorts to a metric-learning-based triplet network to capture accurate, discriminative behavioral representations among IoT devices. Then, inspired by few-shot incremental learning (FSIL), the EAPN further considers advancing incremental model updates based on the relationship between traffic features of new and existing IoT devices. Extensive evaluations demonstrate that the evolvable attention prototypical network (EAPN) not only surpasses state-of-the-art (SOTA) adaptive IoT device-type identification methods but also achieves over 90% accuracy under almost all incremental conditions, maintaining satisfactory performance even after multiple updates.
- 18 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (11) , 24823-24837
Toward Evolvable IoT Device-Type Identification Using Few-Shot Incremental Learning
The proliferation of Internet of Things (IoT) devices has profoundly transformed various industries. However, with the continual emergence of new IoT device types, their extensive heterogeneity and weak security have posed significant challenges for network management and security. Despite existing research excelling at classifying IoT traffic, they have yet to consider incremental updates and timely identification, which are critical for early device management and security in IoT networks. As a solution, we present an evolvable attention prototypical network (EAPN), a novel and evolvable IoT device identification model that supports adapting to new IoT devices with limited traffic. The EAPN extracts traffic features from only a few dozen packets and resorts to a metric-learning-based triplet network to capture accurate, discriminative behavioral representations among IoT devices. Then, inspired by few-shot incremental learning (FSIL), the EAPN further considers advancing incremental model updates based on the relationship between traffic features of new and existing IoT devices. Extensive evaluations demonstrate that the evolvable attention prototypical network (EAPN) not only surpasses state-of-the-art (SOTA) adaptive IoT device-type identification methods but also achieves over 90% accuracy under almost all incremental conditions, maintaining satisfactory performance even after multiple updates.
- 12 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 2252-2267
STMWF: Multi-Tab Website Fingerprinting via Spatial–Temporal Sequence Analysis
Website fingerprinting (WF) attacks are employed to identify websites that utilize Tor encryption. Although State-Of-The-Art (SOTA) WF attacks demonstrate strong performance in single-tab scenarios, they face challenges in multi-tab scenarios. Many multi-tab WF attacks rely solely on direction sequence or process directional and temporal sequence separately. They ignore the coupling between directional and temporal features, which reflects distinct resource-loading processes for different websites. To address the limitations of existing approaches, this paper proposes a new multi-tab WF attack, STMWF. It leverages spatial-temporal sequence analysis and jointly models Inter Arrival Time (IAT) with the direction sequence. STMWF utilizes an SE-attention-based feature extractor to derive features from various website resources within the spatial-temporal sequence. It then employs correlation self-attention to integrate these resource features into their respective websites, ultimately constructing distinct fingerprints for each site. Additionally, the method incorporates correlation denoising to suppress noise in the website fingerprints, thereby enhancing the discriminability of the extracted features. We collected single-tab traces to synthesize a dataset with controlled overlap ratios. We also captured real-world multi-tab traffic with varying tab-opening intervals, evaluating performance under authentic conditions. The experimental results indicate that STMWF significantly outperforms the SOTA multi-tab attacks in both dynamic and static settings. Specifically, it achieves an average F1-score improvement of approximately 14.87% under static conditions and 34.81% under dynamic conditions compared to the SOTA multi-tab WF attack, ARES. Furthermore, STMWF exhibits greater robustness against WF defenses than SOTA attacks and consistently surpasses them across varying overlapping scenarios.
- 12 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 2252-2267
STMWF: Multi-Tab Website Fingerprinting via Spatial–Temporal Sequence Analysis
Website fingerprinting (WF) attacks are employed to identify websites that utilize Tor encryption. Although State-Of-The-Art (SOTA) WF attacks demonstrate strong performance in single-tab scenarios, they face challenges in multi-tab scenarios. Many multi-tab WF attacks rely solely on direction sequence or process directional and temporal sequence separately. They ignore the coupling between directional and temporal features, which reflects distinct resource-loading processes for different websites. To address the limitations of existing approaches, this paper proposes a new multi-tab WF attack, STMWF. It leverages spatial-temporal sequence analysis and jointly models Inter Arrival Time (IAT) with the direction sequence. STMWF utilizes an SE-attention-based feature extractor to derive features from various website resources within the spatial-temporal sequence. It then employs correlation self-attention to integrate these resource features into their respective websites, ultimately constructing distinct fingerprints for each site. Additionally, the method incorporates correlation denoising to suppress noise in the website fingerprints, thereby enhancing the discriminability of the extracted features. We collected single-tab traces to synthesize a dataset with controlled overlap ratios. We also captured real-world multi-tab traffic with varying tab-opening intervals, evaluating performance under authentic conditions. The experimental results indicate that STMWF significantly outperforms the SOTA multi-tab attacks in both dynamic and static settings. Specifically, it achieves an average F1-score improvement of approximately 14.87% under static conditions and 34.81% under dynamic conditions compared to the SOTA multi-tab WF attack, ARES. Furthermore, STMWF exhibits greater robustness against WF defenses than SOTA attacks and consistently surpasses them across varying overlapping scenarios.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 5804-5816
NFTracker: Fine-Grained NFT Behavior Traffic Identification Over Encrypted Tunnel
Non-Fungible Tokens (NFTs) have completely changed digital ownership and the decentralized economy. However, their anonymity and encrypted communication, conducted over encrypted tunnels, pose a significant obstacle to regulating illegal activities. Despite advances in encrypted traffic analysis, fine-grained identification of NFT behaviors over encrypted tunnels faces two critical challenges: 1) inexact segmentation of continuous behavioral traffic, and 2) feature homogeneity due to encryption-induced pattern obfuscation. In this paper, we propose NFTracker, a novel framework to identify fine-grained NFT behavioral traffic over encrypted tunnels. We design a traffic segmentation method to isolate behavioral units by leveraging traffic bursts and distribution discrepancies. To combat feature homogeneity, we introduce a sliding-window-based spatio-temporal feature extraction mechanism that captures localized action fingerprints. Furthermore, we utilize a hybrid CNN-Transformer model to integrate spatial patterns and temporal dependencies for robust behavior identification. We evaluate NFTracker on real-world datasets covering five NFT behaviors (browsing, wallet login, purchasing, selling, and minting). Experimental results demonstrate that NFTracker achieves an average F1-score of 0.9212 on identifying NFT behavioral traffic, outperforming state-of-the-art methods in encrypted tunnel scenarios.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 5804-5816
NFTracker: Fine-Grained NFT Behavior Traffic Identification Over Encrypted Tunnel
Non-Fungible Tokens (NFTs) have completely changed digital ownership and the decentralized economy. However, their anonymity and encrypted communication, conducted over encrypted tunnels, pose a significant obstacle to regulating illegal activities. Despite advances in encrypted traffic analysis, fine-grained identification of NFT behaviors over encrypted tunnels faces two critical challenges: 1) inexact segmentation of continuous behavioral traffic, and 2) feature homogeneity due to encryption-induced pattern obfuscation. In this paper, we propose NFTracker, a novel framework to identify fine-grained NFT behavioral traffic over encrypted tunnels. We design a traffic segmentation method to isolate behavioral units by leveraging traffic bursts and distribution discrepancies. To combat feature homogeneity, we introduce a sliding-window-based spatio-temporal feature extraction mechanism that captures localized action fingerprints. Furthermore, we utilize a hybrid CNN-Transformer model to integrate spatial patterns and temporal dependencies for robust behavior identification. We evaluate NFTracker on real-world datasets covering five NFT behaviors (browsing, wallet login, purchasing, selling, and minting). Experimental results demonstrate that NFTracker achieves an average F1-score of 0.9212 on identifying NFT behavioral traffic, outperforming state-of-the-art methods in encrypted tunnel scenarios.
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-8
NFT-AF: Multi-Dimensional Non-Fungible Token Application Fingerprinting Over Encrypted Tunnels
With the development of the blockchain economy, many criminals are attempting to profit from Non-Fungible Tokens (NFTs). Under the cover of encrypted tunnels, efficiently identifying NFT application traffic has become a challenge. Moreover, NFT applications exhibit a high degree of similarity in communication interfaces, traffic encryption settings, and even business behaviors, making it difficult to distinguish the traffic of different NFT applications. The fact that a regulator can determine the users of a website’s visit by fingerprinting encrypted traffic patterns brings hope for solving this challenge. However, existing Deep Learning (DL) fingerprinting requires complete flows for identification, leading to high training costs and identification latency. This paper presents a novel NFT application fingerprinting method over encrypted tunnels called NFT-AF (NFT Application Fingerprinting). It extracts encrypted tunnel, statistical, and sequential features from early-stage network traffic to generate multi-dimensional fingerprints. Then, it uses a lightweight random forest as the classifier to efficiently identify NFT applications. Our experimental results demonstrate that NFT-AF effectively identifies NFT applications over encrypted tunnels using early-stage session packets and outperforms the state-of-the-art (SOTA) methods.
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-8
NFT-AF: Multi-Dimensional Non-Fungible Token Application Fingerprinting Over Encrypted Tunnels
With the development of the blockchain economy, many criminals are attempting to profit from Non-Fungible Tokens (NFTs). Under the cover of encrypted tunnels, efficiently identifying NFT application traffic has become a challenge. Moreover, NFT applications exhibit a high degree of similarity in communication interfaces, traffic encryption settings, and even business behaviors, making it difficult to distinguish the traffic of different NFT applications. The fact that a regulator can determine the users of a website’s visit by fingerprinting encrypted traffic patterns brings hope for solving this challenge. However, existing Deep Learning (DL) fingerprinting requires complete flows for identification, leading to high training costs and identification latency. This paper presents a novel NFT application fingerprinting method over encrypted tunnels called NFT-AF (NFT Application Fingerprinting). It extracts encrypted tunnel, statistical, and sequential features from early-stage network traffic to generate multi-dimensional fingerprints. Then, it uses a lightweight random forest as the classifier to efficiently identify NFT applications. Our experimental results demonstrate that NFT-AF effectively identifies NFT applications over encrypted tunnels using early-stage session packets and outperforms the state-of-the-art (SOTA) methods.